ml-finance-python

python scripts for finance machine learning
git clone https://9o.is/git/ml-finance-python.git
dfs_quant_finance.ipynb

(476024B)
      1 {
      2  "cells": [
      3   {
      4    "cell_type": "markdown",
      5    "metadata": {},
      6    "source": [
      7     "** Before running this notebook, run the 2 cells containing supporting functions at the bottom. **"
      8    ]
      9   },
     10   {
     11    "cell_type": "markdown",
     12    "metadata": {},
     13    "source": [
     14     "# Daily Fantasy Sports and Quantitative Finance\n",
     15     "### A case study on using quantitative finance tools to build daily fantasy sports lineups.\n",
     16     "\n",
     17     "\n",
     18     "Daily fantasy sports (DFS) have emerged in recent years as a popular skill game. Usually, games challenge participants to pick a team of players (called a 'lineup') in a particular sport on a given day with the objective of getting the most game points on that day. In the classic DFS format, players are assigned a fictional 'salary' or cost and the DFS participant has to pick a roster of players who fit under a total cost constraint, commonly referred to as a 'salary cap'. Players are ranked at the end of the day based on a sport-dependent scoring function.\n",
     19     "\n",
     20     "In this notebook, we will take a look at how quantitative finance tools can be used to study daily fantasy sports data and build an optimal lineup using convex optimization. Specifically, we will take a look at the NBA (basketball) game on DraftKings. The data used in this notebooks is obtained from https://stats.nba.com and [DraftKings](https://www.draftkings.com) (DK). The tools used are a mix of tools from the Quantopian API as well as other Python libraries."
     21    ]
     22   },
     23   {
     24    "cell_type": "markdown",
     25    "metadata": {},
     26    "source": [
     27     "## Load and Format Data\n",
     28     "The first step to most quantitative problems is to get data. In this study, the data was collected and uploaded to the Quantopian research environment as a .csv file. Let's load it into a [pandas DataFrame](http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.html) using [local_csv](https://www.quantopian.com/help#quantopian_research_local_csv) (a Quantopian-specific function)."
     29    ]
     30   },
     31   {
     32    "cell_type": "code",
     33    "execution_count": 2,
     34    "metadata": {},
     35    "outputs": [],
     36    "source": [
     37     "import pandas as pd"
     38    ]
     39   },
     40   {
     41    "cell_type": "code",
     42    "execution_count": 3,
     43    "metadata": {},
     44    "outputs": [],
     45    "source": [
     46     "df = local_csv('nba_data.csv')"
     47    ]
     48   },
     49   {
     50    "cell_type": "code",
     51    "execution_count": 4,
     52    "metadata": {},
     53    "outputs": [],
     54    "source": [
     55     "df['game_datetime'] = pd.to_datetime(df['game_date'])"
     56    ]
     57   },
     58   {
     59    "cell_type": "code",
     60    "execution_count": 5,
     61    "metadata": {},
     62    "outputs": [],
     63    "source": [
     64     "df = df.set_index(['game_datetime', 'player_id'])"
     65    ]
     66   },
     67   {
     68    "cell_type": "markdown",
     69    "metadata": {},
     70    "source": [
     71     "It is important to know what our data looks like, so here is a preview of the first few rows of our DataFrame."
     72    ]
     73   },
     74   {
     75    "cell_type": "code",
     76    "execution_count": 6,
     77    "metadata": {},
     78    "outputs": [
     79     {
     80      "data": {
     81       "text/html": [
     82        "<div>\n",
     83        "<table border=\"1\" class=\"dataframe\">\n",
     84        "  <thead>\n",
     85        "    <tr style=\"text-align: right;\">\n",
     86        "      <th></th>\n",
     87        "      <th></th>\n",
     88        "      <th>game_id</th>\n",
     89        "      <th>season_id</th>\n",
     90        "      <th>game_date</th>\n",
     91        "      <th>matchup</th>\n",
     92        "      <th>wl</th>\n",
     93        "      <th>min</th>\n",
     94        "      <th>fgm</th>\n",
     95        "      <th>fga</th>\n",
     96        "      <th>fg_pct</th>\n",
     97        "      <th>fg3m</th>\n",
     98        "      <th>...</th>\n",
     99        "      <th>height</th>\n",
    100        "      <th>weight</th>\n",
    101        "      <th>birth_date</th>\n",
    102        "      <th>age</th>\n",
    103        "      <th>exp</th>\n",
    104        "      <th>school</th>\n",
    105        "      <th>dk_score</th>\n",
    106        "      <th>dk_salary</th>\n",
    107        "      <th>dk_position</th>\n",
    108        "      <th>dk_position_codes</th>\n",
    109        "    </tr>\n",
    110        "    <tr>\n",
    111        "      <th>game_datetime</th>\n",
    112        "      <th>player_id</th>\n",
    113        "      <th></th>\n",
    114        "      <th></th>\n",
    115        "      <th></th>\n",
    116        "      <th></th>\n",
    117        "      <th></th>\n",
    118        "      <th></th>\n",
    119        "      <th></th>\n",
    120        "      <th></th>\n",
    121        "      <th></th>\n",
    122        "      <th></th>\n",
    123        "      <th></th>\n",
    124        "      <th></th>\n",
    125        "      <th></th>\n",
    126        "      <th></th>\n",
    127        "      <th></th>\n",
    128        "      <th></th>\n",
    129        "      <th></th>\n",
    130        "      <th></th>\n",
    131        "      <th></th>\n",
    132        "      <th></th>\n",
    133        "      <th></th>\n",
    134        "    </tr>\n",
    135        "  </thead>\n",
    136        "  <tbody>\n",
    137        "    <tr>\n",
    138        "      <th>2018-04-11</th>\n",
    139        "      <th>1713</th>\n",
    140        "      <td>21701230</td>\n",
    141        "      <td>22017</td>\n",
    142        "      <td>2018-04-11</td>\n",
    143        "      <td>SAC vs. HOU</td>\n",
    144        "      <td>W</td>\n",
    145        "      <td>25</td>\n",
    146        "      <td>5</td>\n",
    147        "      <td>8</td>\n",
    148        "      <td>0.625</td>\n",
    149        "      <td>2</td>\n",
    150        "      <td>...</td>\n",
    151        "      <td>6-6</td>\n",
    152        "      <td>220</td>\n",
    153        "      <td>JAN 26, 1977</td>\n",
    154        "      <td>41.0</td>\n",
    155        "      <td>19</td>\n",
    156        "      <td>North Carolina</td>\n",
    157        "      <td>17.75</td>\n",
    158        "      <td>NaN</td>\n",
    159        "      <td>NaN</td>\n",
    160        "      <td>NaN</td>\n",
    161        "    </tr>\n",
    162        "    <tr>\n",
    163        "      <th>2018-04-09</th>\n",
    164        "      <th>1713</th>\n",
    165        "      <td>21701210</td>\n",
    166        "      <td>22017</td>\n",
    167        "      <td>2018-04-09</td>\n",
    168        "      <td>SAC @ SAS</td>\n",
    169        "      <td>L</td>\n",
    170        "      <td>17</td>\n",
    171        "      <td>1</td>\n",
    172        "      <td>2</td>\n",
    173        "      <td>0.500</td>\n",
    174        "      <td>1</td>\n",
    175        "      <td>...</td>\n",
    176        "      <td>6-6</td>\n",
    177        "      <td>220</td>\n",
    178        "      <td>JAN 26, 1977</td>\n",
    179        "      <td>41.0</td>\n",
    180        "      <td>19</td>\n",
    181        "      <td>North Carolina</td>\n",
    182        "      <td>14.50</td>\n",
    183        "      <td>NaN</td>\n",
    184        "      <td>NaN</td>\n",
    185        "      <td>NaN</td>\n",
    186        "    </tr>\n",
    187        "    <tr>\n",
    188        "      <th>2018-04-06</th>\n",
    189        "      <th>1713</th>\n",
    190        "      <td>21701188</td>\n",
    191        "      <td>22017</td>\n",
    192        "      <td>2018-04-06</td>\n",
    193        "      <td>SAC @ MEM</td>\n",
    194        "      <td>W</td>\n",
    195        "      <td>13</td>\n",
    196        "      <td>0</td>\n",
    197        "      <td>4</td>\n",
    198        "      <td>0.000</td>\n",
    199        "      <td>0</td>\n",
    200        "      <td>...</td>\n",
    201        "      <td>6-6</td>\n",
    202        "      <td>220</td>\n",
    203        "      <td>JAN 26, 1977</td>\n",
    204        "      <td>41.0</td>\n",
    205        "      <td>19</td>\n",
    206        "      <td>North Carolina</td>\n",
    207        "      <td>9.75</td>\n",
    208        "      <td>3400.0</td>\n",
    209        "      <td>SG/SF</td>\n",
    210        "      <td>[0, 1, 0, 1, 0, 1, 1, 1]</td>\n",
    211        "    </tr>\n",
    212        "  </tbody>\n",
    213        "</table>\n",
    214        "<p>3 rows × 42 columns</p>\n",
    215        "</div>"
    216       ],
    217       "text/plain": [
    218        "                          game_id  season_id   game_date      matchup wl  min  \\\n",
    219        "game_datetime player_id                                                         \n",
    220        "2018-04-11    1713       21701230      22017  2018-04-11  SAC vs. HOU  W   25   \n",
    221        "2018-04-09    1713       21701210      22017  2018-04-09    SAC @ SAS  L   17   \n",
    222        "2018-04-06    1713       21701188      22017  2018-04-06    SAC @ MEM  W   13   \n",
    223        "\n",
    224        "                         fgm  fga  fg_pct  fg3m            ...             \\\n",
    225        "game_datetime player_id                                    ...              \n",
    226        "2018-04-11    1713         5    8   0.625     2            ...              \n",
    227        "2018-04-09    1713         1    2   0.500     1            ...              \n",
    228        "2018-04-06    1713         0    4   0.000     0            ...              \n",
    229        "\n",
    230        "                         height  weight    birth_date   age  exp  \\\n",
    231        "game_datetime player_id                                            \n",
    232        "2018-04-11    1713          6-6     220  JAN 26, 1977  41.0   19   \n",
    233        "2018-04-09    1713          6-6     220  JAN 26, 1977  41.0   19   \n",
    234        "2018-04-06    1713          6-6     220  JAN 26, 1977  41.0   19   \n",
    235        "\n",
    236        "                                 school  dk_score  dk_salary  dk_position  \\\n",
    237        "game_datetime player_id                                                     \n",
    238        "2018-04-11    1713       North Carolina     17.75        NaN          NaN   \n",
    239        "2018-04-09    1713       North Carolina     14.50        NaN          NaN   \n",
    240        "2018-04-06    1713       North Carolina      9.75     3400.0        SG/SF   \n",
    241        "\n",
    242        "                                dk_position_codes  \n",
    243        "game_datetime player_id                            \n",
    244        "2018-04-11    1713                            NaN  \n",
    245        "2018-04-09    1713                            NaN  \n",
    246        "2018-04-06    1713       [0, 1, 0, 1, 0, 1, 1, 1]  \n",
    247        "\n",
    248        "[3 rows x 42 columns]"
    249       ]
    250      },
    251      "execution_count": 6,
    252      "metadata": {},
    253      "output_type": "execute_result"
    254     }
    255    ],
    256    "source": [
    257     "df.head(3)"
    258    ]
    259   },
    260   {
    261    "cell_type": "markdown",
    262    "metadata": {},
    263    "source": [
    264     "And here is a look at a single row where we can see all of the columns."
    265    ]
    266   },
    267   {
    268    "cell_type": "code",
    269    "execution_count": 7,
    270    "metadata": {},
    271    "outputs": [
    272     {
    273      "data": {
    274       "text/plain": [
    275        "game_id                              21701057\n",
    276        "season_id                               22017\n",
    277        "game_date                          2018-03-19\n",
    278        "matchup                           SAC vs. DET\n",
    279        "wl                                          L\n",
    280        "min                                        17\n",
    281        "fgm                                         3\n",
    282        "fga                                         6\n",
    283        "fg_pct                                    0.5\n",
    284        "fg3m                                        1\n",
    285        "fg3a                                        4\n",
    286        "fg3_pct                                  0.25\n",
    287        "ftm                                         0\n",
    288        "fta                                         0\n",
    289        "ft_pct                                      0\n",
    290        "oreb                                        1\n",
    291        "dreb                                        0\n",
    292        "reb                                         1\n",
    293        "ast                                         0\n",
    294        "stl                                         2\n",
    295        "blk                                         1\n",
    296        "tov                                         2\n",
    297        "pf                                          3\n",
    298        "pts                                         7\n",
    299        "plus_minus                                 -3\n",
    300        "video_available                             1\n",
    301        "team_id                            1610612758\n",
    302        "season                                   2017\n",
    303        "league_id                                   0\n",
    304        "player                           Vince Carter\n",
    305        "num                                        15\n",
    306        "position                                  G-F\n",
    307        "height                                    6-6\n",
    308        "weight                                    220\n",
    309        "birth_date                       JAN 26, 1977\n",
    310        "age                                        41\n",
    311        "exp                                        19\n",
    312        "school                         North Carolina\n",
    313        "dk_score                                13.75\n",
    314        "dk_salary                                3300\n",
    315        "dk_position                             SG/SF\n",
    316        "dk_position_codes    [0, 1, 0, 1, 0, 1, 1, 1]\n",
    317        "Name: (2018-03-19 00:00:00, 1713), dtype: object"
    318       ]
    319      },
    320      "execution_count": 7,
    321      "metadata": {},
    322      "output_type": "execute_result"
    323     }
    324    ],
    325    "source": [
    326     "df.iloc[10]"
    327    ]
    328   },
    329   {
    330    "cell_type": "markdown",
    331    "metadata": {},
    332    "source": [
    333     "## Alphalens (Factor Analysis)\n",
    334     "[Alphalens](https://github.com/quantopian/alphalens) is a tool on Quantopian for analyzing the predictve ability of a factor. In this notebook, we will use Alphalens to analyze the predictive ability of a **fantasy points per game** factor.\n",
    335     "\n",
    336     "The first step to analzing our factor is to define it. Here, we will define the fantasy points per game factor, `trailing_20_dk_score` to be the rolling mean fantasy points per game (using DK scoring rules) over the last 20 days."
    337    ]
    338   },
    339   {
    340    "cell_type": "code",
    341    "execution_count": 8,
    342    "metadata": {},
    343    "outputs": [],
    344    "source": [
    345     "# Rolling 20 day mean fantasy score per game.\n",
    346     "trailing_20_dk_score = df['dk_score'].unstack().shift(1).rolling(20, min_periods=5).mean().stack()"
    347    ]
    348   },
    349   {
    350    "cell_type": "code",
    351    "execution_count": 14,
    352    "metadata": {},
    353    "outputs": [
    354     {
    355      "data": {
    356       "text/plain": [
    357        "game_datetime  player_id\n",
    358        "2017-10-26     1717         20.10\n",
    359        "               2037         20.70\n",
    360        "               2199         22.40\n",
    361        "               2207         15.50\n",
    362        "               2544         55.95\n",
    363        "dtype: float64"
    364       ]
    365      },
    366      "execution_count": 14,
    367      "metadata": {},
    368      "output_type": "execute_result"
    369     }
    370    ],
    371    "source": [
    372     "trailing_20_dk_score.head()"
    373    ]
    374   },
    375   {
    376    "cell_type": "markdown",
    377    "metadata": {},
    378    "source": [
    379     "And then we need to define our objective metric, or 'what we want to maximize'. In finance, we want to maximize portfolio returns. In DraftKings DFS, we want to maximize our DraftKings score for the current day."
    380    ]
    381   },
    382   {
    383    "cell_type": "code",
    384    "execution_count": 11,
    385    "metadata": {},
    386    "outputs": [],
    387    "source": [
    388     "# DK scores.\n",
    389     "scores = df['dk_score'].unstack()"
    390    ]
    391   },
    392   {
    393    "cell_type": "markdown",
    394    "metadata": {},
    395    "source": [
    396     "Next, we need to import Alphalens so that we can use it to analyze our factors."
    397    ]
    398   },
    399   {
    400    "cell_type": "code",
    401    "execution_count": 12,
    402    "metadata": {},
    403    "outputs": [],
    404    "source": [
    405     "import alphalens as al"
    406    ]
    407   },
    408   {
    409    "cell_type": "markdown",
    410    "metadata": {},
    411    "source": [
    412     "Before using Alphalens, we need to format our data in a way that it expects. Let's start with our points factor. The below function creates a DataFrame using our points factor and our DK scores. It also specifies `periods` to be 1, meaning we only care about the performance of the player on one day. The data will be categorized into 5 'quantiles' to remove some of the noise from the relationship."
    413    ]
    414   },
    415   {
    416    "cell_type": "markdown",
    417    "metadata": {},
    418    "source": [
    419     "Now, let's plot the relationship between each of our points per game factor quantiles and see how strongly they predict DK points on a given day."
    420    ]
    421   },
    422   {
    423    "cell_type": "code",
    424    "execution_count": 13,
    425    "metadata": {},
    426    "outputs": [],
    427    "source": [
    428     "# Format rolling fantasy points per game data for use in Alphalens.\n",
    429     "factor_data = get_clean_factor_and_forward_scores(\n",
    430     "    factor=trailing_20_dk_score,\n",
    431     "    scores=scores.loc['2017-11-09':],\n",
    432     "    periods=[1],\n",
    433     ")"
    434    ]
    435   },
    436   {
    437    "cell_type": "code",
    438    "execution_count": 14,
    439    "metadata": {},
    440    "outputs": [],
    441    "source": [
    442     "mean_by_q_daily, std_err_by_q_daily = al.performance.mean_return_by_quantile(factor_data,\n",
    443     "                                                                                       by_date=True,\n",
    444     "                                                                                       demeaned=False)"
    445    ]
    446   },
    447   {
    448    "cell_type": "code",
    449    "execution_count": 15,
    450    "metadata": {},
    451    "outputs": [
    452     {
    453      "data": {
    454       "image/png": "iVBORw0KGgoAAAANSUhEUgAABBwAAAF8CAYAAABsV6j4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
    455       "text/plain": [
    456        "<matplotlib.figure.Figure at 0x7f7a55848b90>"
    457       ]
    458      },
    459      "metadata": {},
    460      "output_type": "display_data"
    461     }
    462    ],
    463    "source": [
    464     "al.plotting.plot_quantile_returns_violin(mean_by_q_daily/10000);"
    465    ]
    466   },
    467   {
    468    "cell_type": "markdown",
    469    "metadata": {},
    470    "source": [
    471     "It looks like our fantasy points per game factor is a strong predictor. We see a  narrow and distinct distribution around each quantile's mean DK points (y axis).\n",
    472     "\n",
    473     "Let's look at another visualization of the points per game factor and plot out the daily mean of each quantile."
    474    ]
    475   },
    476   {
    477    "cell_type": "code",
    478    "execution_count": 16,
    479    "metadata": {},
    480    "outputs": [],
    481    "source": [
    482     "quant_spread, std_err_spread = al.performance.compute_mean_returns_spread(mean_by_q_daily,\n",
    483     "                                                                                 upper_quant=5,\n",
    484     "                                                                                 lower_quant=1,\n",
    485     "                                                                                 std_err=std_err_by_q_daily)"
    486    ]
    487   },
    488   {
    489    "cell_type": "code",
    490    "execution_count": 17,
    491    "metadata": {},
    492    "outputs": [
    493     {
    494      "data": {
    495       "text/plain": [
    496        "<matplotlib.text.Text at 0x7f7a398f2a90>"
    497       ]
    498      },
    499      "execution_count": 17,
    500      "metadata": {},
    501      "output_type": "execute_result"
    502     },
    503     {
    504      "data": {
    505       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzwAAAH6CAYAAADC2EluAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
    506       "text/plain": [
    507        "<matplotlib.figure.Figure at 0x7f7a51798290>"
    508       ]
    509      },
    510      "metadata": {},
    511      "output_type": "display_data"
    512     }
    513    ],
    514    "source": [
    515     "import matplotlib.cm as cm\n",
    516     "import matplotlib.pyplot as plt\n",
    517     "\n",
    518     "plt.figure(figsize=(14,8))\n",
    519     "quantiles = mean_by_q_daily.index.get_level_values(0)\n",
    520     "colors = cm.rainbow(np.linspace(0, 1, len(quantiles.unique())))\n",
    521     "for quantile in quantiles.unique():\n",
    522     "    x = mean_by_q_daily[quantiles == quantile].index.get_level_values(1)\n",
    523     "    y = mean_by_q_daily[quantiles == quantile]\n",
    524     "    plt.plot(x, y, color=colors[quantile-1])\n",
    525     "    \n",
    526     "ax = plt.gca()\n",
    527     "ax.legend(quantiles.unique())\n",
    528     "ax.set_ylabel('DK Points')\n",
    529     "ax.set_xlabel('Date')\n",
    530     "\n",
    531     "plt.title('Daily Mean DK Points Per Quantile (Points Per Game Factor)')"
    532    ]
    533   },
    534   {
    535    "cell_type": "markdown",
    536    "metadata": {},
    537    "source": [
    538     "## Optimize (Portfolio Optimization)\n",
    539     "Now that we have identified some strong predictors of fantasy points, let's use the data to build an optimal portfolio each day. Specifically, we will pick a team of players (construct a portfolio) based on their mean points per game over the last 20 days.\n",
    540     "\n",
    541     "On Quantopian, we can use the [Optimize API](https://www.quantopian.com/help#optimize-title) to help us calculate an optimal 'portfolio' each day. In finance, this means picking the stocks that we should hold each day based on an objective function and a set of constraints. In daily fantasy sports, the problem is the same: we want to pick a lineup based on an objective function (maximize DK points) subject to a series of constraints (based on the rules of the game).\n",
    542     "\n",
    543     "Generally speaking, constraints are the easy part. The rules are set before every game and we know that our lineup has to follow these rules. Specifically, the NBA game at DraftKings has the following [rules](https://www.draftkings.com/help/rules/4):\n",
    544     "- The lineup must be composed of exactly 8 players.\n",
    545     "- The sum of the fictional player salaries in a lineup must be less than 50,000.\n",
    546     "- A lineup must include players from at least two NBA games.\n",
    547     "- A lineup must include at least one player classified in each of the following positions:\n",
    548     "    - PG\n",
    549     "    - SG\n",
    550     "    - PF\n",
    551     "    - SF\n",
    552     "    - C\n",
    553     "    - G (PG or SG, in addition to the other PG and SG slots)\n",
    554     "    - F (PF or SF, in addition to the other PF and SF slots)\n",
    555     "    - U (the last spot can be occupied by a player of any position).\n",
    556     "    \n",
    557     "The challenging part is creating an objective function. We know that we want to maximize the number of DK points that we get on a given day, but without time travel, it's difficult to know exactly what the function looks like. In practice, this means we need to define an 'expected value' objective that represents our expected performance of each player. Earlier, we saw that our trailing points per game factor was a good predictor of DK points, so let's use that as our objective function.\n",
    558     "\n",
    559     "As an optimization problem, this means that we want to pick a team of players subject to the DK rules (salary cap, positions, etc.) that maximizes the total trailing points per game factor.\n",
    560     "\n",
    561     "Let's define our problem using the Optimize API."
    562    ]
    563   },
    564   {
    565    "cell_type": "code",
    566    "execution_count": 18,
    567    "metadata": {},
    568    "outputs": [],
    569    "source": [
    570     "# Import the Optimize namespace.\n",
    571     "import quantopian.optimize as opt"
    572    ]
    573   },
    574   {
    575    "cell_type": "markdown",
    576    "metadata": {},
    577    "source": [
    578     "Let's start by assembling and formatting the player salaries and positions so that we can use them as constraints in our optimization problem."
    579    ]
    580   },
    581   {
    582    "cell_type": "code",
    583    "execution_count": 19,
    584    "metadata": {},
    585    "outputs": [],
    586    "source": [
    587     "# Player costs.\n",
    588     "player_costs = pd.DataFrame(df['dk_salary']).dropna().sort_index(level=['game_datetime', 'player_id'])\n",
    589     "\n",
    590     "# Player positions.\n",
    591     "player_positions = df['dk_position_codes'].dropna().apply((lambda x: [int(y) for y in x[1:-1].split(',')]))"
    592    ]
    593   },
    594   {
    595    "cell_type": "markdown",
    596    "metadata": {},
    597    "source": [
    598     "Then, let's restructure our (`trailing_20_dk_score`) so that we can use it as an objective function in Optimize. We'll define it to be parametric on the date so that we can play around with the date later on."
    599    ]
    600   },
    601   {
    602    "cell_type": "code",
    603    "execution_count": 20,
    604    "metadata": {},
    605    "outputs": [],
    606    "source": [
    607     "# Define the game date that we want to test. Change this and re-run \n",
    608     "# cells below to build a lineup for another date.\n",
    609     "TEST_DATE = '2017-12-01'"
    610    ]
    611   },
    612   {
    613    "cell_type": "code",
    614    "execution_count": 21,
    615    "metadata": {},
    616    "outputs": [],
    617    "source": [
    618     "# Format our expected value factor (trailing_20_dk_score) for use in Optimize.\n",
    619     "expected_value = trailing_20_dk_score[player_costs.index].dropna()\n",
    620     "expected_value = expected_value[expected_value.index.get_level_values(0) == TEST_DATE]"
    621    ]
    622   },
    623   {
    624    "cell_type": "markdown",
    625    "metadata": {},
    626    "source": [
    627     "### Objective\n",
    628     "Next, let's define our objective function using [optimize.MaximizeAlpha](https://www.quantopian.com/help#quantopian_optimize_MaximizeAlpha), which will try to maximize the value of our lineup based on our `expected_value` factor."
    629    ]
    630   },
    631   {
    632    "cell_type": "code",
    633    "execution_count": 22,
    634    "metadata": {},
    635    "outputs": [],
    636    "source": [
    637     "# Objective function to be fed into Optimize later on.\n",
    638     "obj = opt.MaximizeAlpha(expected_value)"
    639    ]
    640   },
    641   {
    642    "cell_type": "markdown",
    643    "metadata": {},
    644    "source": [
    645     "### Constraints\n",
    646     "Many of the built-in constraints are centered around finance problems, so we will have to make some simplifying assumptions in order to use them to build our lineup. Later in the notebook, we will build our own optimizer to build a lineup that strictly follows the DK rules.\n",
    647     "\n",
    648     "Let's start by defining our salary cap constraint:"
    649    ]
    650   },
    651   {
    652    "cell_type": "code",
    653    "execution_count": 23,
    654    "metadata": {},
    655    "outputs": [],
    656    "source": [
    657     "# Salary cap constraint. The total cost of our lineup has to be less than 50,000.\n",
    658     "# We will define this as a FactorExposure constraint.\n",
    659     "cost_limit = opt.FactorExposure(\n",
    660     "    player_costs, \n",
    661     "    min_exposures={'dk_salary': 0}, \n",
    662     "    max_exposures={'dk_salary': 50000}\n",
    663     ")"
    664    ]
    665   },
    666   {
    667    "cell_type": "markdown",
    668    "metadata": {},
    669    "source": [
    670     "Next, we'll define a position constraint. To simplify things, we'll use more generic positions, and limit ourselves to a maximum of 4 forwards (`F`), 4 guards (`G`), and 2 centers (`C`), with a minimum of 3 `F`, 3 `G`, and 1 `C`."
    671    ]
    672   },
    673   {
    674    "cell_type": "code",
    675    "execution_count": 24,
    676    "metadata": {},
    677    "outputs": [],
    678    "source": [
    679     "# Map from each player to its position.\n",
    680     "labels = df['position'].apply(lambda x: x[0])\n",
    681     "\n",
    682     "# # Maps from each position to its min/max exposure.\n",
    683     "min_exposures = {'F': 3, 'G': 3, 'C': 1}\n",
    684     "max_exposures = {'F': 4, 'G': 4, 'C': 2}\n",
    685     "\n",
    686     "player_position_constraint = opt.NetGroupExposure(labels, min_exposures, max_exposures)"
    687    ]
    688   },
    689   {
    690    "cell_type": "markdown",
    691    "metadata": {},
    692    "source": [
    693     "At this point, it's worth noting that Optimize operates in weight space, meaning it chooses a percentage (not bounded by a total of 100%, can also be positive or negative) of the overall lineup to allocate to each asset/player. In an attempt to get around that, the next constraint we define will specify that we can hold no more than 1 of each player, and no fewer than 0 of each player (you can't [short sell](https://www.investopedia.com/terms/s/shortselling.asp) a player... yet)."
    694    ]
    695   },
    696   {
    697    "cell_type": "code",
    698    "execution_count": 25,
    699    "metadata": {},
    700    "outputs": [],
    701    "source": [
    702     "# This constraints tells the Optimizer than we can hold at most 1 of each player\n",
    703     "discretize_players = opt.PositionConcentration(\n",
    704     "    pd.Series([]),\n",
    705     "    pd.Series([]),\n",
    706     "    default_min_weight=0,\n",
    707     "    default_max_weight=1,\n",
    708     ")"
    709    ]
    710   },
    711   {
    712    "cell_type": "markdown",
    713    "metadata": {},
    714    "source": [
    715     "Lastly, let's define a constraint for our total number of players."
    716    ]
    717   },
    718   {
    719    "cell_type": "code",
    720    "execution_count": 26,
    721    "metadata": {},
    722    "outputs": [],
    723    "source": [
    724     "max_players = opt.MaxGrossExposure(8)"
    725    ]
    726   },
    727   {
    728    "cell_type": "markdown",
    729    "metadata": {},
    730    "source": [
    731     "### Calculate Optimal Portfolio (Build Lineup)\n",
    732     "The next step is to calculate our optimal portfolio for the day using our objective and constraints."
    733    ]
    734   },
    735   {
    736    "cell_type": "code",
    737    "execution_count": 27,
    738    "metadata": {},
    739    "outputs": [],
    740    "source": [
    741     "result = opt.calculate_optimal_portfolio(\n",
    742     "    objective=obj,\n",
    743     "    constraints=[\n",
    744     "        discretize_players,\n",
    745     "        player_position_constraint,\n",
    746     "        max_players,\n",
    747     "        cost_limit,\n",
    748     "    ]\n",
    749     ")"
    750    ]
    751   },
    752   {
    753    "cell_type": "markdown",
    754    "metadata": {},
    755    "source": [
    756     "And here are the resulting weights:"
    757    ]
    758   },
    759   {
    760    "cell_type": "code",
    761    "execution_count": 28,
    762    "metadata": {},
    763    "outputs": [],
    764    "source": [
    765     "resulting_picks = result[(result.index.get_level_values(0) == TEST_DATE)]\n",
    766     "player_weights = resulting_picks[resulting_picks>0]"
    767    ]
    768   },
    769   {
    770    "cell_type": "code",
    771    "execution_count": 29,
    772    "metadata": {},
    773    "outputs": [],
    774    "source": [
    775     "lineup_info = df.loc[player_weights.index][['matchup', 'player', 'position', 'dk_salary']]"
    776    ]
    777   },
    778   {
    779    "cell_type": "code",
    780    "execution_count": 30,
    781    "metadata": {},
    782    "outputs": [],
    783    "source": [
    784     "lineup_info['weight'] = player_weights"
    785    ]
    786   },
    787   {
    788    "cell_type": "code",
    789    "execution_count": 31,
    790    "metadata": {},
    791    "outputs": [
    792     {
    793      "data": {
    794       "text/html": [
    795        "<div>\n",
    796        "<table border=\"1\" class=\"dataframe\">\n",
    797        "  <thead>\n",
    798        "    <tr style=\"text-align: right;\">\n",
    799        "      <th></th>\n",
    800        "      <th></th>\n",
    801        "      <th>matchup</th>\n",
    802        "      <th>player</th>\n",
    803        "      <th>position</th>\n",
    804        "      <th>dk_salary</th>\n",
    805        "      <th>weight</th>\n",
    806        "    </tr>\n",
    807        "    <tr>\n",
    808        "      <th>game_datetime</th>\n",
    809        "      <th>player_id</th>\n",
    810        "      <th></th>\n",
    811        "      <th></th>\n",
    812        "      <th></th>\n",
    813        "      <th></th>\n",
    814        "      <th></th>\n",
    815        "    </tr>\n",
    816        "  </thead>\n",
    817        "  <tbody>\n",
    818        "    <tr>\n",
    819        "      <th rowspan=\"9\" valign=\"top\">2017-12-01</th>\n",
    820        "      <th>200768</th>\n",
    821        "      <td>TOR vs. IND</td>\n",
    822        "      <td>Kyle Lowry</td>\n",
    823        "      <td>G</td>\n",
    824        "      <td>8100.0</td>\n",
    825        "      <td>1.000000</td>\n",
    826        "    </tr>\n",
    827        "    <tr>\n",
    828        "      <th>201188</th>\n",
    829        "      <td>MEM vs. SAS</td>\n",
    830        "      <td>Marc Gasol</td>\n",
    831        "      <td>C</td>\n",
    832        "      <td>7800.0</td>\n",
    833        "      <td>1.000000</td>\n",
    834        "    </tr>\n",
    835        "    <tr>\n",
    836        "      <th>201988</th>\n",
    837        "      <td>SAS @ MEM</td>\n",
    838        "      <td>Patty Mills</td>\n",
    839        "      <td>G</td>\n",
    840        "      <td>4100.0</td>\n",
    841        "      <td>1.000000</td>\n",
    842        "    </tr>\n",
    843        "    <tr>\n",
    844        "      <th>202331</th>\n",
    845        "      <td>OKC vs. MIN</td>\n",
    846        "      <td>Paul George</td>\n",
    847        "      <td>F</td>\n",
    848        "      <td>8200.0</td>\n",
    849        "      <td>1.000000</td>\n",
    850        "    </tr>\n",
    851        "    <tr>\n",
    852        "      <th>202692</th>\n",
    853        "      <td>UTA vs. NOP</td>\n",
    854        "      <td>Alec Burks</td>\n",
    855        "      <td>G</td>\n",
    856        "      <td>3600.0</td>\n",
    857        "      <td>0.459459</td>\n",
    858        "    </tr>\n",
    859        "    <tr>\n",
    860        "      <th>202710</th>\n",
    861        "      <td>MIN @ OKC</td>\n",
    862        "      <td>Jimmy Butler</td>\n",
    863        "      <td>G</td>\n",
    864        "      <td>7300.0</td>\n",
    865        "      <td>0.540541</td>\n",
    866        "    </tr>\n",
    867        "    <tr>\n",
    868        "      <th>202711</th>\n",
    869        "      <td>IND @ TOR</td>\n",
    870        "      <td>Bojan Bogdanovic</td>\n",
    871        "      <td>F</td>\n",
    872        "      <td>4600.0</td>\n",
    873        "      <td>1.000000</td>\n",
    874        "    </tr>\n",
    875        "    <tr>\n",
    876        "      <th>202734</th>\n",
    877        "      <td>NOP @ UTA</td>\n",
    878        "      <td>E'Twaun Moore</td>\n",
    879        "      <td>F-G</td>\n",
    880        "      <td>3600.0</td>\n",
    881        "      <td>1.000000</td>\n",
    882        "    </tr>\n",
    883        "    <tr>\n",
    884        "      <th>203506</th>\n",
    885        "      <td>IND @ TOR</td>\n",
    886        "      <td>Victor Oladipo</td>\n",
    887        "      <td>G</td>\n",
    888        "      <td>8000.0</td>\n",
    889        "      <td>1.000000</td>\n",
    890        "    </tr>\n",
    891        "  </tbody>\n",
    892        "</table>\n",
    893        "</div>"
    894       ],
    895       "text/plain": [
    896        "                             matchup            player position  dk_salary  \\\n",
    897        "game_datetime player_id                                                      \n",
    898        "2017-12-01    200768     TOR vs. IND        Kyle Lowry        G     8100.0   \n",
    899        "              201188     MEM vs. SAS        Marc Gasol        C     7800.0   \n",
    900        "              201988       SAS @ MEM       Patty Mills        G     4100.0   \n",
    901        "              202331     OKC vs. MIN       Paul George        F     8200.0   \n",
    902        "              202692     UTA vs. NOP        Alec Burks        G     3600.0   \n",
    903        "              202710       MIN @ OKC      Jimmy Butler        G     7300.0   \n",
    904        "              202711       IND @ TOR  Bojan Bogdanovic        F     4600.0   \n",
    905        "              202734       NOP @ UTA     E'Twaun Moore      F-G     3600.0   \n",
    906        "              203506       IND @ TOR    Victor Oladipo        G     8000.0   \n",
    907        "\n",
    908        "                           weight  \n",
    909        "game_datetime player_id            \n",
    910        "2017-12-01    200768     1.000000  \n",
    911        "              201188     1.000000  \n",
    912        "              201988     1.000000  \n",
    913        "              202331     1.000000  \n",
    914        "              202692     0.459459  \n",
    915        "              202710     0.540541  \n",
    916        "              202711     1.000000  \n",
    917        "              202734     1.000000  \n",
    918        "              203506     1.000000  "
    919       ]
    920      },
    921      "execution_count": 31,
    922      "metadata": {},
    923      "output_type": "execute_result"
    924     }
    925    ],
    926    "source": [
    927     "lineup_info"
    928    ]
    929   },
    930   {
    931    "cell_type": "markdown",
    932    "metadata": {},
    933    "source": [
    934     "Something that immediately pops out is the non-binary weight assigned to Kemba Walker. Obviously, it's not possible to pick 0.55% of a player in DFS, so this lineup won't be eligible to enter. Unfortunately, it's not currently possible to binarize the outputs of Optimize on Quantopian (since it's not a helpful feature in quantitative finance).\n",
    935     "\n",
    936     "# CVXPY\n",
    937     "To get around the limits of Optimize, we can take a similar approach and use [cvxpy](http://www.cvxpy.org/en/latest/), a Python library for solving convex optimization problems (and the library on top of which Optimize was built), to build our own objective and constraints. In this version, we will follow the exact rules of the DraftKings NBA game (8 players, \\$50k salary cap, players in at least two games, and one player in each of the following positions: ['PG', 'SG', 'PF', 'SF', 'C', 'G', 'F', 'U']. The following cells build constraints to enforce these rules."
    938    ]
    939   },
    940   {
    941    "cell_type": "code",
    942    "execution_count": 32,
    943    "metadata": {},
    944    "outputs": [],
    945    "source": [
    946     "# List of DK positions.\n",
    947     "DK_POSITION_LIST = ['PG', 'SG', 'PF', 'SF', 'C', 'G', 'F', 'U']"
    948    ]
    949   },
    950   {
    951    "cell_type": "code",
    952    "execution_count": 36,
    953    "metadata": {},
    954    "outputs": [],
    955    "source": [
    956     "# Get the average play time for all players over the last 20 games. We will use this to\n",
    957     "# filter out players with skewed stats due to playing very little.\n",
    958     "trailing_20_mins = df['min'].unstack().shift(1).rolling(20, min_periods=5).mean().stack()"
    959    ]
    960   },
    961   {
    962    "cell_type": "code",
    963    "execution_count": 37,
    964    "metadata": {},
    965    "outputs": [],
    966    "source": [
    967     "# Players with non-null salary on TEST_DATE.\n",
    968     "have_salary = player_costs[player_costs.index.get_level_values(0) == TEST_DATE].dropna().index\n",
    969     "\n",
    970     "# Values from our factor for TEST_DATE.\n",
    971     "trailing_dk_score_today = trailing_20_dk_score[\n",
    972     "    (trailing_20_dk_score.index.get_level_values(0) == TEST_DATE)\n",
    973     "]\n",
    974     "\n",
    975     "# Players with at least 5 minutes per game on average over the last 20 days.\n",
    976     "trailing_mins_today = trailing_20_mins[(trailing_20_mins.index.get_level_values(0) == TEST_DATE)]\n",
    977     "have_play_time = trailing_mins_today[trailing_mins_today >= 5].index\n",
    978     "\n",
    979     "# Eligible players for us to pick have a non-null salary and at least 5 minutes\n",
    980     "# played per game over the last 20 days.\n",
    981     "eligible_players_today = have_salary.intersection(have_play_time)\n",
    982     "\n",
    983     "# The game ID in which each player is playing.\n",
    984     "player_games = df.loc[eligible_players_today].game_id\n",
    985     "\n",
    986     "# The set of all game IDs on TEST_DATE\n",
    987     "todays_games = df.loc[eligible_players_today].game_id.unique().tolist()"
    988    ]
    989   },
    990   {
    991    "cell_type": "markdown",
    992    "metadata": {},
    993    "source": [
    994     "Next, we will import cvxpy and define our objective and constraints."
    995    ]
    996   },
    997   {
    998    "cell_type": "code",
    999    "execution_count": 38,
   1000    "metadata": {},
   1001    "outputs": [],
   1002    "source": [
   1003     "import cvxpy as cvx"
   1004    ]
   1005   },
   1006   {
   1007    "cell_type": "code",
   1008    "execution_count": 39,
   1009    "metadata": {
   1010     "scrolled": true
   1011    },
   1012    "outputs": [
   1013     {
   1014      "name": "stdout",
   1015      "output_type": "stream",
   1016      "text": [
   1017       "Our total expected value from today's lineup is:\n"
   1018      ]
   1019     },
   1020     {
   1021      "data": {
   1022       "text/plain": [
   1023        "273.37121208187256"
   1024       ]
   1025      },
   1026      "execution_count": 39,
   1027      "metadata": {},
   1028      "output_type": "execute_result"
   1029     }
   1030    ],
   1031    "source": [
   1032     "# Player salaries and expected values.\n",
   1033     "salaries = np.squeeze(np.array(player_costs.loc[eligible_players_today]))\n",
   1034     "values = np.array(expected_value[eligible_players_today])\n",
   1035     "\n",
   1036     "# The variable we are solving for. We define our output variable as a Bool\n",
   1037     "# since we have to make a binary decision on each player (pick or don't pick).\n",
   1038     "selection = cvx.Bool(len(salaries))\n",
   1039     "selection.is_positive = new_is_positive\n",
   1040     "\n",
   1041     "# Our lineup's total salary must be less than 50,000.\n",
   1042     "salary_cap = 50000\n",
   1043     "cost_constraint = salaries * selection <= salary_cap\n",
   1044     "\n",
   1045     "# Our lineup must be composed of exactly 8 players.\n",
   1046     "player_constraint = np.ones(len(salaries)) * selection == 8\n",
   1047     "\n",
   1048     "# Our total expected value is the sum of the value of each player in\n",
   1049     "# the lineup. We define our objective to maximize the total expected\n",
   1050     "# value.\n",
   1051     "total_expected_value = values * selection\n",
   1052     "objective = cvx.Maximize(total_expected_value)\n",
   1053     "\n",
   1054     "# Put our cost and player count constraints in a list.\n",
   1055     "constraints = [cost_constraint, player_constraint]\n",
   1056     "\n",
   1057     "# Define our position constraints. Positions are represented along an 8-element\n",
   1058     "# array corresponding to the positions in DK_POSITION_LIST.\n",
   1059     "position_min = np.array([1, 1, 1, 1, 1, 3, 3, 8])\n",
   1060     "pos_limits = {}\n",
   1061     "i = 0\n",
   1062     "for pos in DK_POSITION_LIST:\n",
   1063     "    pos_limits[pos] = np.array(player_positions[eligible_players_today].apply(lambda x: x[i]))\n",
   1064     "    constraints.append((pos_limits[pos] * selection) >= position_min[i])\n",
   1065     "    i += 1\n",
   1066     "\n",
   1067     "# Define our game constraints. We rephrase the rule as 'you cannot pick more than\n",
   1068     "# 7 players from any one game'.\n",
   1069     "for gid in todays_games:\n",
   1070     "    game_limit = np.array(player_games == gid)\n",
   1071     "    constraints.append((game_limit * selection) <= 7)\n",
   1072     "    \n",
   1073     "    \n",
   1074     "\n",
   1075     "# We tell cvxpy that we want maximize our expected value, subject to all\n",
   1076     "# of our constraints.\n",
   1077     "optimization_problem = cvx.Problem(objective, constraints)\n",
   1078     "\n",
   1079     "print \"Our total expected value from today's lineup is:\"\n",
   1080     "\n",
   1081     "# Solving the problem.\n",
   1082     "optimization_problem.solve(solver=cvx.ECOS_BB)"
   1083    ]
   1084   },
   1085   {
   1086    "cell_type": "markdown",
   1087    "metadata": {},
   1088    "source": [
   1089     "Based on our player-by-player expected values and the constraints we supplied to `cvxpy`, our optimal lineup has a total expected value of 273.37.\n",
   1090     "\n",
   1091     "Let's take a look at who is in this lineup to make sure that we implemented the rules properly:"
   1092    ]
   1093   },
   1094   {
   1095    "cell_type": "code",
   1096    "execution_count": 40,
   1097    "metadata": {},
   1098    "outputs": [],
   1099    "source": [
   1100     "# Format output and get relevant player info for display.\n",
   1101     "dk_team = pd.Series(np.squeeze(selection.value).tolist()[0])\n",
   1102     "player_info = df.loc[player_costs.loc[eligible_players_today].iloc[dk_team[dk_team > 0.1].index.values].index][['matchup', 'player', 'dk_position', 'dk_salary', 'dk_score']]"
   1103    ]
   1104   },
   1105   {
   1106    "cell_type": "code",
   1107    "execution_count": 41,
   1108    "metadata": {},
   1109    "outputs": [
   1110     {
   1111      "data": {
   1112       "text/html": [
   1113        "<div>\n",
   1114        "<table border=\"1\" class=\"dataframe\">\n",
   1115        "  <thead>\n",
   1116        "    <tr style=\"text-align: right;\">\n",
   1117        "      <th></th>\n",
   1118        "      <th></th>\n",
   1119        "      <th>matchup</th>\n",
   1120        "      <th>player</th>\n",
   1121        "      <th>dk_position</th>\n",
   1122        "      <th>dk_salary</th>\n",
   1123        "      <th>dk_score</th>\n",
   1124        "    </tr>\n",
   1125        "    <tr>\n",
   1126        "      <th>game_datetime</th>\n",
   1127        "      <th>player_id</th>\n",
   1128        "      <th></th>\n",
   1129        "      <th></th>\n",
   1130        "      <th></th>\n",
   1131        "      <th></th>\n",
   1132        "      <th></th>\n",
   1133        "    </tr>\n",
   1134        "  </thead>\n",
   1135        "  <tbody>\n",
   1136        "    <tr>\n",
   1137        "      <th rowspan=\"8\" valign=\"top\">2017-12-01</th>\n",
   1138        "      <th>200768</th>\n",
   1139        "      <td>TOR vs. IND</td>\n",
   1140        "      <td>Kyle Lowry</td>\n",
   1141        "      <td>PG</td>\n",
   1142        "      <td>8100.0</td>\n",
   1143        "      <td>31.75</td>\n",
   1144        "    </tr>\n",
   1145        "    <tr>\n",
   1146        "      <th>201959</th>\n",
   1147        "      <td>MIN @ OKC</td>\n",
   1148        "      <td>Taj Gibson</td>\n",
   1149        "      <td>PF/C</td>\n",
   1150        "      <td>5700.0</td>\n",
   1151        "      <td>16.00</td>\n",
   1152        "    </tr>\n",
   1153        "    <tr>\n",
   1154        "      <th>202331</th>\n",
   1155        "      <td>OKC vs. MIN</td>\n",
   1156        "      <td>Paul George</td>\n",
   1157        "      <td>SF</td>\n",
   1158        "      <td>8200.0</td>\n",
   1159        "      <td>62.00</td>\n",
   1160        "    </tr>\n",
   1161        "    <tr>\n",
   1162        "      <th>202692</th>\n",
   1163        "      <td>UTA vs. NOP</td>\n",
   1164        "      <td>Alec Burks</td>\n",
   1165        "      <td>PG/SG</td>\n",
   1166        "      <td>3600.0</td>\n",
   1167        "      <td>35.75</td>\n",
   1168        "    </tr>\n",
   1169        "    <tr>\n",
   1170        "      <th>202710</th>\n",
   1171        "      <td>MIN @ OKC</td>\n",
   1172        "      <td>Jimmy Butler</td>\n",
   1173        "      <td>SG</td>\n",
   1174        "      <td>7300.0</td>\n",
   1175        "      <td>45.25</td>\n",
   1176        "    </tr>\n",
   1177        "    <tr>\n",
   1178        "      <th>203087</th>\n",
   1179        "      <td>CHA @ MIA</td>\n",
   1180        "      <td>Jeremy Lamb</td>\n",
   1181        "      <td>SG/SF</td>\n",
   1182        "      <td>5700.0</td>\n",
   1183        "      <td>24.50</td>\n",
   1184        "    </tr>\n",
   1185        "    <tr>\n",
   1186        "      <th>203506</th>\n",
   1187        "      <td>IND @ TOR</td>\n",
   1188        "      <td>Victor Oladipo</td>\n",
   1189        "      <td>PG/SG</td>\n",
   1190        "      <td>8000.0</td>\n",
   1191        "      <td>66.25</td>\n",
   1192        "    </tr>\n",
   1193        "    <tr>\n",
   1194        "      <th>1627832</th>\n",
   1195        "      <td>TOR vs. IND</td>\n",
   1196        "      <td>Fred VanVleet</td>\n",
   1197        "      <td>PG</td>\n",
   1198        "      <td>3400.0</td>\n",
   1199        "      <td>27.50</td>\n",
   1200        "    </tr>\n",
   1201        "  </tbody>\n",
   1202        "</table>\n",
   1203        "</div>"
   1204       ],
   1205       "text/plain": [
   1206        "                             matchup          player dk_position  dk_salary  \\\n",
   1207        "game_datetime player_id                                                       \n",
   1208        "2017-12-01    200768     TOR vs. IND      Kyle Lowry          PG     8100.0   \n",
   1209        "              201959       MIN @ OKC      Taj Gibson        PF/C     5700.0   \n",
   1210        "              202331     OKC vs. MIN     Paul George          SF     8200.0   \n",
   1211        "              202692     UTA vs. NOP      Alec Burks       PG/SG     3600.0   \n",
   1212        "              202710       MIN @ OKC    Jimmy Butler          SG     7300.0   \n",
   1213        "              203087       CHA @ MIA     Jeremy Lamb       SG/SF     5700.0   \n",
   1214        "              203506       IND @ TOR  Victor Oladipo       PG/SG     8000.0   \n",
   1215        "              1627832    TOR vs. IND   Fred VanVleet          PG     3400.0   \n",
   1216        "\n",
   1217        "                         dk_score  \n",
   1218        "game_datetime player_id            \n",
   1219        "2017-12-01    200768        31.75  \n",
   1220        "              201959        16.00  \n",
   1221        "              202331        62.00  \n",
   1222        "              202692        35.75  \n",
   1223        "              202710        45.25  \n",
   1224        "              203087        24.50  \n",
   1225        "              203506        66.25  \n",
   1226        "              1627832       27.50  "
   1227       ]
   1228      },
   1229      "execution_count": 41,
   1230      "metadata": {},
   1231      "output_type": "execute_result"
   1232     }
   1233    ],
   1234    "source": [
   1235     "player_info"
   1236    ]
   1237   },
   1238   {
   1239    "cell_type": "code",
   1240    "execution_count": 48,
   1241    "metadata": {},
   1242    "outputs": [
   1243     {
   1244      "name": "stdout",
   1245      "output_type": "stream",
   1246      "text": [
   1247       "Total lineup salary: 50000\n",
   1248       "Total actual score: 309\n"
   1249      ]
   1250     }
   1251    ],
   1252    "source": [
   1253     "print \"Total lineup salary: %d\" % player_info['dk_salary'].sum()\n",
   1254     "print \"Total actual score: %d\" % player_info['dk_score'].sum()"
   1255    ]
   1256   },
   1257   {
   1258    "cell_type": "markdown",
   1259    "metadata": {},
   1260    "source": [
   1261     "Our roster appears to satisfy the rules!"
   1262    ]
   1263   },
   1264   {
   1265    "cell_type": "markdown",
   1266    "metadata": {},
   1267    "source": [
   1268     "# Backtest (a.k.a. run the above code over several consecutive days)"
   1269    ]
   1270   },
   1271   {
   1272    "cell_type": "code",
   1273    "execution_count": 42,
   1274    "metadata": {},
   1275    "outputs": [],
   1276    "source": [
   1277     "def backtest(factor, dates, filters, df, player_costs, player_positions):\n",
   1278     "    historical_results = {}\n",
   1279     "    \n",
   1280     "    daily_expected_value = factor[filters].dropna()\n",
   1281     "    \n",
   1282     "    for _date in dates:\n",
   1283     "        try:\n",
   1284     "            print _date\n",
   1285     "            daily_filter = filters[filters.get_level_values(0) == _date]\n",
   1286     "            expected_value_today = daily_expected_value[daily_expected_value.index.get_level_values(0) == _date]\n",
   1287     "\n",
   1288     "            # The game ID in which each player is playing.\n",
   1289     "            player_games = df.loc[daily_filter].game_id\n",
   1290     "\n",
   1291     "            # The set of all game IDs on the current date.\n",
   1292     "            todays_games = df.loc[daily_filter].game_id.unique().tolist()\n",
   1293     "\n",
   1294     "            # Player salaries and expected values.\n",
   1295     "            salaries = np.squeeze(np.array(player_costs.loc[daily_filter]))\n",
   1296     "            values = np.array(expected_value_today[daily_filter])\n",
   1297     "\n",
   1298     "            # The variable we are solving for. We define our output variable as a Bool\n",
   1299     "            # since we have to make a binary decision on each player (pick or don't pick).\n",
   1300     "            selection = cvx.Bool(len(salaries))\n",
   1301     "            selection.is_positive = new_is_positive\n",
   1302     "\n",
   1303     "            # Our lineup's total salary must be less than 50,000.\n",
   1304     "            salary_cap = 50000\n",
   1305     "            cost_constraint = salaries * selection <= salary_cap\n",
   1306     "\n",
   1307     "            # Our lineup must be composed of exactly 8 players.\n",
   1308     "            player_constraint = np.ones(len(salaries)) * selection == 8\n",
   1309     "\n",
   1310     "            # Our total expected value is the sum of the value of each player in\n",
   1311     "            # the lineup. We define our objective to maximize the total expected\n",
   1312     "            # value.\n",
   1313     "            total_expected_value = values * selection\n",
   1314     "            objective = cvx.Maximize(total_expected_value)\n",
   1315     "\n",
   1316     "            # Put our cost and player count constraints in a list.\n",
   1317     "            constraints = [cost_constraint, player_constraint]\n",
   1318     "\n",
   1319     "            # Define our position constraints. Positions are represented along an 8-element\n",
   1320     "            # array corresponding to the positions in DK_POSITION_LIST.\n",
   1321     "            position_min = np.array([1, 1, 1, 1, 1, 3, 3, 8])\n",
   1322     "            pos_limits = {}\n",
   1323     "            i = 0\n",
   1324     "            for pos in DK_POSITION_LIST:\n",
   1325     "                pos_limits[pos] = np.array(player_positions[daily_filter].apply(lambda x: x[i]))\n",
   1326     "                constraints.append((pos_limits[pos] * selection) >= position_min[i])\n",
   1327     "                i += 1\n",
   1328     "\n",
   1329     "            # Define our game constraints. We rephrase the rule as 'you cannot pick more than\n",
   1330     "            # 7 players from any one game'.\n",
   1331     "            for gid in todays_games:\n",
   1332     "                game_limit = np.array(player_games == gid)\n",
   1333     "                constraints.append((game_limit * selection) <= 7)\n",
   1334     "\n",
   1335     "\n",
   1336     "            # We tell cvxpy that we want maximize our expected value, subject to all\n",
   1337     "            # of our constraints.\n",
   1338     "            knapsack_problem = cvx.Problem(objective, constraints)\n",
   1339     "\n",
   1340     "            # Solving the problem.\n",
   1341     "            predicted_value = knapsack_problem.solve(solver=cvx.ECOS_BB)\n",
   1342     "\n",
   1343     "            dk_team = pd.Series(np.squeeze(selection.value).tolist()[0])\n",
   1344     "            player_info = df.loc[player_costs.loc[filters].iloc[dk_team[dk_team > 0.1].index.values].index][['matchup', 'player', 'dk_position', 'dk_salary', 'dk_score']]\n",
   1345     "            historical_results[_date] = {\n",
   1346     "                'actual_score': player_info.dk_score.sum(),\n",
   1347     "                'expected_score': predicted_value,\n",
   1348     "            }\n",
   1349     "        except TypeError:\n",
   1350     "            pass\n",
   1351     "        except ValueError:\n",
   1352     "            pass\n",
   1353     "    \n",
   1354     "    return historical_results"
   1355    ]
   1356   },
   1357   {
   1358    "cell_type": "code",
   1359    "execution_count": 45,
   1360    "metadata": {},
   1361    "outputs": [],
   1362    "source": [
   1363     "test_dates = trailing_20_dk_score.index.get_level_values(0).unique().values[15:]\n",
   1364     "has_playtime = trailing_20_mins[trailing_20_mins >= 5]"
   1365    ]
   1366   },
   1367   {
   1368    "cell_type": "code",
   1369    "execution_count": 46,
   1370    "metadata": {
   1371     "scrolled": true
   1372    },
   1373    "outputs": [
   1374     {
   1375      "name": "stdout",
   1376      "output_type": "stream",
   1377      "text": [
   1378       "2017-11-10T00:00:00.000000000\n",
   1379       "2017-11-11T00:00:00.000000000\n",
   1380       "2017-11-12T00:00:00.000000000\n",
   1381       "2017-11-13T00:00:00.000000000\n",
   1382       "2017-11-14T00:00:00.000000000\n",
   1383       "2017-11-15T00:00:00.000000000\n",
   1384       "2017-11-16T00:00:00.000000000\n",
   1385       "2017-11-17T00:00:00.000000000\n",
   1386       "2017-11-18T00:00:00.000000000\n",
   1387       "2017-11-19T00:00:00.000000000\n",
   1388       "2017-11-20T00:00:00.000000000\n",
   1389       "2017-11-21T00:00:00.000000000\n",
   1390       "2017-11-22T00:00:00.000000000\n",
   1391       "2017-11-24T00:00:00.000000000\n",
   1392       "2017-11-25T00:00:00.000000000\n",
   1393       "2017-11-26T00:00:00.000000000\n",
   1394       "2017-11-27T00:00:00.000000000\n",
   1395       "2017-11-28T00:00:00.000000000\n",
   1396       "2017-11-29T00:00:00.000000000\n",
   1397       "2017-11-30T00:00:00.000000000\n",
   1398       "2017-12-01T00:00:00.000000000\n",
   1399       "2017-12-02T00:00:00.000000000\n",
   1400       "2017-12-03T00:00:00.000000000\n",
   1401       "2017-12-04T00:00:00.000000000\n",
   1402       "2017-12-05T00:00:00.000000000\n",
   1403       "2017-12-06T00:00:00.000000000\n",
   1404       "2017-12-07T00:00:00.000000000\n",
   1405       "2017-12-08T00:00:00.000000000\n",
   1406       "2017-12-09T00:00:00.000000000\n",
   1407       "2017-12-10T00:00:00.000000000\n",
   1408       "2017-12-11T00:00:00.000000000\n",
   1409       "2017-12-12T00:00:00.000000000\n",
   1410       "2017-12-13T00:00:00.000000000\n",
   1411       "2017-12-14T00:00:00.000000000\n",
   1412       "2017-12-15T00:00:00.000000000\n",
   1413       "2017-12-16T00:00:00.000000000\n",
   1414       "2017-12-17T00:00:00.000000000\n",
   1415       "2017-12-18T00:00:00.000000000\n",
   1416       "2017-12-19T00:00:00.000000000\n",
   1417       "2017-12-20T00:00:00.000000000\n",
   1418       "2017-12-21T00:00:00.000000000\n",
   1419       "2017-12-22T00:00:00.000000000\n",
   1420       "2017-12-23T00:00:00.000000000\n",
   1421       "2017-12-25T00:00:00.000000000\n",
   1422       "2017-12-26T00:00:00.000000000\n",
   1423       "2017-12-27T00:00:00.000000000\n",
   1424       "2017-12-28T00:00:00.000000000\n",
   1425       "2017-12-29T00:00:00.000000000\n",
   1426       "2017-12-30T00:00:00.000000000\n",
   1427       "2017-12-31T00:00:00.000000000\n",
   1428       "2018-01-01T00:00:00.000000000\n",
   1429       "2018-01-02T00:00:00.000000000\n",
   1430       "2018-01-03T00:00:00.000000000\n",
   1431       "2018-01-04T00:00:00.000000000\n",
   1432       "2018-01-05T00:00:00.000000000\n",
   1433       "2018-01-06T00:00:00.000000000\n",
   1434       "2018-01-07T00:00:00.000000000\n",
   1435       "2018-01-08T00:00:00.000000000\n",
   1436       "2018-01-09T00:00:00.000000000\n",
   1437       "2018-01-10T00:00:00.000000000\n",
   1438       "2018-01-11T00:00:00.000000000\n",
   1439       "2018-01-12T00:00:00.000000000\n",
   1440       "2018-01-13T00:00:00.000000000\n",
   1441       "2018-01-14T00:00:00.000000000\n",
   1442       "2018-01-15T00:00:00.000000000\n",
   1443       "2018-01-16T00:00:00.000000000\n",
   1444       "2018-01-17T00:00:00.000000000\n",
   1445       "2018-01-18T00:00:00.000000000\n",
   1446       "2018-01-19T00:00:00.000000000\n",
   1447       "2018-01-20T00:00:00.000000000\n",
   1448       "2018-01-21T00:00:00.000000000\n",
   1449       "2018-01-22T00:00:00.000000000\n",
   1450       "2018-01-23T00:00:00.000000000\n",
   1451       "2018-01-24T00:00:00.000000000\n",
   1452       "2018-01-25T00:00:00.000000000\n",
   1453       "2018-01-26T00:00:00.000000000\n",
   1454       "2018-01-27T00:00:00.000000000\n",
   1455       "2018-01-28T00:00:00.000000000\n",
   1456       "2018-01-29T00:00:00.000000000\n",
   1457       "2018-01-30T00:00:00.000000000\n",
   1458       "2018-01-31T00:00:00.000000000\n",
   1459       "2018-02-01T00:00:00.000000000\n",
   1460       "2018-02-02T00:00:00.000000000\n",
   1461       "2018-02-03T00:00:00.000000000\n",
   1462       "2018-02-04T00:00:00.000000000\n",
   1463       "2018-02-05T00:00:00.000000000\n",
   1464       "2018-02-06T00:00:00.000000000\n",
   1465       "2018-02-07T00:00:00.000000000\n",
   1466       "2018-02-08T00:00:00.000000000\n",
   1467       "2018-02-09T00:00:00.000000000\n",
   1468       "2018-02-10T00:00:00.000000000\n",
   1469       "2018-02-11T00:00:00.000000000\n",
   1470       "2018-02-12T00:00:00.000000000\n",
   1471       "2018-02-13T00:00:00.000000000\n",
   1472       "2018-02-14T00:00:00.000000000\n",
   1473       "2018-02-15T00:00:00.000000000\n",
   1474       "2018-02-22T00:00:00.000000000\n",
   1475       "2018-02-23T00:00:00.000000000\n",
   1476       "2018-02-24T00:00:00.000000000\n",
   1477       "2018-02-25T00:00:00.000000000\n",
   1478       "2018-02-26T00:00:00.000000000\n",
   1479       "2018-02-27T00:00:00.000000000\n",
   1480       "2018-02-28T00:00:00.000000000\n",
   1481       "2018-03-01T00:00:00.000000000\n",
   1482       "2018-03-02T00:00:00.000000000\n",
   1483       "2018-03-03T00:00:00.000000000\n",
   1484       "2018-03-04T00:00:00.000000000\n",
   1485       "2018-03-05T00:00:00.000000000\n",
   1486       "2018-03-06T00:00:00.000000000\n",
   1487       "2018-03-07T00:00:00.000000000\n",
   1488       "2018-03-08T00:00:00.000000000\n",
   1489       "2018-03-09T00:00:00.000000000\n",
   1490       "2018-03-10T00:00:00.000000000\n",
   1491       "2018-03-11T00:00:00.000000000\n",
   1492       "2018-03-12T00:00:00.000000000\n",
   1493       "2018-03-13T00:00:00.000000000\n",
   1494       "2018-03-14T00:00:00.000000000\n",
   1495       "2018-03-15T00:00:00.000000000\n",
   1496       "2018-03-16T00:00:00.000000000\n",
   1497       "2018-03-17T00:00:00.000000000\n",
   1498       "2018-03-18T00:00:00.000000000\n",
   1499       "2018-03-19T00:00:00.000000000\n",
   1500       "2018-03-20T00:00:00.000000000\n",
   1501       "2018-03-21T00:00:00.000000000\n",
   1502       "2018-03-22T00:00:00.000000000\n",
   1503       "2018-03-23T00:00:00.000000000\n",
   1504       "2018-03-24T00:00:00.000000000\n",
   1505       "2018-03-25T00:00:00.000000000\n",
   1506       "2018-03-26T00:00:00.000000000\n",
   1507       "2018-03-27T00:00:00.000000000\n",
   1508       "2018-03-28T00:00:00.000000000\n",
   1509       "2018-03-29T00:00:00.000000000\n",
   1510       "2018-03-30T00:00:00.000000000\n",
   1511       "2018-03-31T00:00:00.000000000\n",
   1512       "2018-04-01T00:00:00.000000000\n",
   1513       "2018-04-03T00:00:00.000000000\n",
   1514       "2018-04-04T00:00:00.000000000\n",
   1515       "2018-04-05T00:00:00.000000000\n",
   1516       "2018-04-06T00:00:00.000000000\n",
   1517       "2018-04-07T00:00:00.000000000\n",
   1518       "2018-04-08T00:00:00.000000000\n",
   1519       "2018-04-09T00:00:00.000000000\n",
   1520       "2018-04-10T00:00:00.000000000\n",
   1521       "2018-04-11T00:00:00.000000000\n"
   1522      ]
   1523     }
   1524    ],
   1525    "source": [
   1526     "filters = player_costs.index & has_playtime.index\n",
   1527     "backtest_result = backtest(\n",
   1528     "    trailing_20_dk_score,\n",
   1529     "    test_dates,\n",
   1530     "    filters,\n",
   1531     "    df,\n",
   1532     "    player_costs,\n",
   1533     "    player_positions,\n",
   1534     ")"
   1535    ]
   1536   },
   1537   {
   1538    "cell_type": "code",
   1539    "execution_count": 47,
   1540    "metadata": {},
   1541    "outputs": [
   1542     {
   1543      "name": "stdout",
   1544      "output_type": "stream",
   1545      "text": [
   1546       "actual_score      189.060714\n",
   1547       "expected_score    276.071713\n",
   1548       "dtype: float64\n"
   1549      ]
   1550     },
   1551     {
   1552      "data": {
   1553       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzMAAAHSCAYAAADYLJW4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngl4SyYwgCIIgCIIgCHrp/wD2UV/IlbUugwAAAABJRU5ErkJggg==\n",
   1554       "text/plain": [
   1555        "<matplotlib.figure.Figure at 0x7f7a39942390>"
   1556       ]
   1557      },
   1558      "metadata": {},
   1559      "output_type": "display_data"
   1560     }
   1561    ],
   1562    "source": [
   1563     "daily_results = pd.DataFrame.from_dict(backtest_result, orient='index')\n",
   1564     "print daily_results.mean()\n",
   1565     "daily_results.plot();"
   1566    ]
   1567   },
   1568   {
   1569    "cell_type": "markdown",
   1570    "metadata": {},
   1571    "source": [
   1572     "## Conclusion\n",
   1573     "In this notebook, we looked at daily fantasy sports using quantitative finance tools. We started out by pulling in and looking at NBA data. We created and tested the ability of certain statistics to predict DK points using Alphalens. We attempted to build an optimal lineup using Optimize. And lastly, we used cvxpy to build a proper solution to the daily lineup construction problem.\n",
   1574     "\n",
   1575     "## Future Work\n",
   1576     "In both quantitative finance and daily fantasy sports, the hardest problem is forecasting results of an asset or player in the future. Now that we have a notebook built to help us test factors and build lineups, the next step is to iterate on ideas of predictive signals. In this example, we only looked at two relatively simple metrics (minutes played and points scored). It would be interesting to try new factors to see if they are predictive or provide a leg up on the competitiong.\n",
   1577     "\n",
   1578     "Another area that we didn't explore in this notebook is risk. In most team sports, the measured statistics of one player tend to be correlated (positively or negatively) with other players in the game. For example, players might get fantasy points for getting an assist when a teammate scores. These correlations can be defined as risk factors. Risk factors can be exploited or hedged depending on the desired risk profile of the lineup. The objective function or constraints in the optimization problem could be altered to either try to exploit or hedge risk."
   1579    ]
   1580   },
   1581   {
   1582    "cell_type": "code",
   1583    "execution_count": null,
   1584    "metadata": {
   1585     "collapsed": true
   1586    },
   1587    "outputs": [],
   1588    "source": []
   1589   },
   1590   {
   1591    "cell_type": "code",
   1592    "execution_count": null,
   1593    "metadata": {
   1594     "collapsed": true
   1595    },
   1596    "outputs": [],
   1597    "source": []
   1598   },
   1599   {
   1600    "cell_type": "code",
   1601    "execution_count": null,
   1602    "metadata": {
   1603     "collapsed": true
   1604    },
   1605    "outputs": [],
   1606    "source": []
   1607   },
   1608   {
   1609    "cell_type": "code",
   1610    "execution_count": null,
   1611    "metadata": {
   1612     "collapsed": true
   1613    },
   1614    "outputs": [],
   1615    "source": []
   1616   },
   1617   {
   1618    "cell_type": "code",
   1619    "execution_count": null,
   1620    "metadata": {
   1621     "collapsed": true
   1622    },
   1623    "outputs": [],
   1624    "source": []
   1625   },
   1626   {
   1627    "cell_type": "code",
   1628    "execution_count": null,
   1629    "metadata": {
   1630     "collapsed": true
   1631    },
   1632    "outputs": [],
   1633    "source": []
   1634   },
   1635   {
   1636    "cell_type": "code",
   1637    "execution_count": null,
   1638    "metadata": {
   1639     "collapsed": true
   1640    },
   1641    "outputs": [],
   1642    "source": []
   1643   },
   1644   {
   1645    "cell_type": "code",
   1646    "execution_count": null,
   1647    "metadata": {
   1648     "collapsed": true
   1649    },
   1650    "outputs": [],
   1651    "source": []
   1652   },
   1653   {
   1654    "cell_type": "code",
   1655    "execution_count": null,
   1656    "metadata": {
   1657     "collapsed": true
   1658    },
   1659    "outputs": [],
   1660    "source": []
   1661   },
   1662   {
   1663    "cell_type": "code",
   1664    "execution_count": null,
   1665    "metadata": {
   1666     "collapsed": true
   1667    },
   1668    "outputs": [],
   1669    "source": []
   1670   },
   1671   {
   1672    "cell_type": "code",
   1673    "execution_count": null,
   1674    "metadata": {
   1675     "collapsed": true
   1676    },
   1677    "outputs": [],
   1678    "source": []
   1679   },
   1680   {
   1681    "cell_type": "markdown",
   1682    "metadata": {},
   1683    "source": [
   1684     "# Helper Functions - Cells Below Need to Be Run at Notebook Start"
   1685    ]
   1686   },
   1687   {
   1688    "cell_type": "code",
   1689    "execution_count": 15,
   1690    "metadata": {},
   1691    "outputs": [],
   1692    "source": [
   1693     "import numpy as np\n",
   1694     "from pandas.tseries.offsets import CustomBusinessDay\n",
   1695     "\n",
   1696     "from scipy.stats import mode\n",
   1697     "\n",
   1698     "def compute_forward_scores(factor,\n",
   1699     "                            scores,\n",
   1700     "                            periods=(1, 5, 10),\n",
   1701     "                            filter_zscore=None):\n",
   1702     "    \"\"\"\n",
   1703     "    Finds the N period forward returns (as percent change) for each asset\n",
   1704     "    provided.\n",
   1705     "    Parameters\n",
   1706     "    ----------\n",
   1707     "    factor : pd.Series - MultiIndex\n",
   1708     "        A MultiIndex Series indexed by timestamp (level 0) and asset\n",
   1709     "        (level 1), containing the values for a single alpha factor.\n",
   1710     "        - See full explanation in utils.get_clean_factor_and_forward_returns\n",
   1711     "    scores : pd.DataFrame\n",
   1712     "        Pricing data to use in forward price calculation.\n",
   1713     "        Assets as columns, dates as index. Pricing data must\n",
   1714     "        span the factor analysis time period plus an additional buffer window\n",
   1715     "        that is greater than the maximum number of expected periods\n",
   1716     "        in the forward returns calculations.\n",
   1717     "    periods : sequence[int]\n",
   1718     "        periods to compute forward returns on.\n",
   1719     "    filter_zscore : int or float, optional\n",
   1720     "        Sets forward returns greater than X standard deviations\n",
   1721     "        from the the mean to nan. Set it to 'None' to avoid filtering.\n",
   1722     "        Caution: this outlier filtering incorporates lookahead bias.\n",
   1723     "    Returns\n",
   1724     "    -------\n",
   1725     "    forward_returns : pd.DataFrame - MultiIndex\n",
   1726     "        A MultiIndex DataFrame indexed by timestamp (level 0) and asset\n",
   1727     "        (level 1), containing the forward returns for assets.\n",
   1728     "        Forward returns column names follow the format accepted by\n",
   1729     "        pd.Timedelta (e.g. '1D', '30m', '3h15m', '1D1h', etc).\n",
   1730     "        'date' index freq property (forward_returns.index.levels[0].freq)\n",
   1731     "        will be set to a trading calendar (pandas DateOffset) inferred\n",
   1732     "        from the input data (see infer_trading_calendar for more details).\n",
   1733     "    \"\"\"\n",
   1734     "\n",
   1735     "    factor_dateindex = factor.index.levels[0]\n",
   1736     "    if factor_dateindex.tz != scores.index.tz:\n",
   1737     "        raise NonMatchingTimezoneError(\"The timezone of 'factor' is not the \"\n",
   1738     "                                       \"same as the timezone of 'scores'. See \"\n",
   1739     "                                       \"the pandas methods tz_localize and \"\n",
   1740     "                                       \"tz_convert.\")\n",
   1741     "\n",
   1742     "    freq = infer_trading_calendar(factor_dateindex, scores.index)\n",
   1743     "\n",
   1744     "    factor_dateindex = factor_dateindex.intersection(scores.index)\n",
   1745     "\n",
   1746     "    if len(factor_dateindex) == 0:\n",
   1747     "        raise ValueError(\"Factor and scores indices don't match: make sure \"\n",
   1748     "                         \"they have the same convention in terms of datetimes \"\n",
   1749     "                         \"and symbol-names\")\n",
   1750     "\n",
   1751     "    forward_returns = pd.DataFrame(index=pd.MultiIndex.from_product(\n",
   1752     "        [factor_dateindex, scores.columns], names=['date', 'asset']))\n",
   1753     "\n",
   1754     "    forward_returns.index.levels[0].freq = freq\n",
   1755     "\n",
   1756     "    for period in sorted(periods):\n",
   1757     "        #\n",
   1758     "        # build forward returns\n",
   1759     "        #\n",
   1760     "        fwdret = (scores\n",
   1761     "                  #.shift(-period)\n",
   1762     "                  .reindex(factor_dateindex)\n",
   1763     "                  )\n",
   1764     "\n",
   1765     "        if filter_zscore is not None:\n",
   1766     "            mask = abs(fwdret - fwdret.mean()) > (filter_zscore * fwdret.std())\n",
   1767     "            fwdret[mask] = np.nan\n",
   1768     "\n",
   1769     "        # Find the period length, which will be the column name\n",
   1770     "        # Becase the calendar inferred from factor and prices doesn't take\n",
   1771     "        # into consideration holidays yet, there could be some non-trading days\n",
   1772     "        # in between the trades so we'll test several entries to find out the\n",
   1773     "        # correct period length\n",
   1774     "        #\n",
   1775     "        entries_to_test = min(10, len(fwdret.index), len(scores.index)-period)\n",
   1776     "        days_diffs = []\n",
   1777     "        for i in range(entries_to_test):\n",
   1778     "            p_idx = scores.index.get_loc(fwdret.index[i])\n",
   1779     "            start = scores.index[p_idx]\n",
   1780     "            end = scores.index[p_idx+period]\n",
   1781     "            period_len = diff_custom_calendar_timedeltas(start, end, freq)\n",
   1782     "            days_diffs.append(period_len.components.days)\n",
   1783     "\n",
   1784     "        delta_days = period_len.components.days - mode(days_diffs).mode[0]\n",
   1785     "        period_len -= pd.Timedelta(days=delta_days)\n",
   1786     "\n",
   1787     "        # Finally use period_len as column name\n",
   1788     "        column_name = timedelta_to_string(period_len)\n",
   1789     "        forward_returns[column_name] = fwdret.stack()\n",
   1790     "\n",
   1791     "    forward_returns.index = forward_returns.index.rename(['date', 'asset'])\n",
   1792     "\n",
   1793     "    return forward_returns\n",
   1794     "\n",
   1795     "def get_clean_factor_and_forward_scores(factor,\n",
   1796     "                                         scores,\n",
   1797     "                                         groupby=None,\n",
   1798     "                                         binning_by_group=False,\n",
   1799     "                                         quantiles=5,\n",
   1800     "                                         bins=None,\n",
   1801     "                                         periods=(1, 5, 10),\n",
   1802     "                                         filter_zscore=20,\n",
   1803     "                                         groupby_labels=None,\n",
   1804     "                                         max_loss=10.0):\n",
   1805     "    \n",
   1806     "    forward_scores = compute_forward_scores(factor, scores, periods,\n",
   1807     "                                              filter_zscore)\n",
   1808     "\n",
   1809     "    factor_data = get_clean_factor(factor, forward_scores, groupby=groupby,\n",
   1810     "                                   groupby_labels=groupby_labels,\n",
   1811     "                                   quantiles=quantiles, bins=bins,\n",
   1812     "                                   binning_by_group=binning_by_group,\n",
   1813     "                                   max_loss=max_loss)\n",
   1814     "\n",
   1815     "    return factor_data\n",
   1816     "\n",
   1817     "\n",
   1818     "def infer_trading_calendar(factor_idx, prices_idx):\n",
   1819     "    \"\"\"\n",
   1820     "    Infer the trading calendar from factor and price information.\n",
   1821     "    Parameters\n",
   1822     "    ----------\n",
   1823     "    factor_idx : pd.DatetimeIndex\n",
   1824     "        The factor datetimes for which we are computing the forward returns\n",
   1825     "    prices_idx : pd.DatetimeIndex\n",
   1826     "        The prices datetimes associated withthe factor data\n",
   1827     "    Returns\n",
   1828     "    -------\n",
   1829     "    calendar : pd.DateOffset\n",
   1830     "    \"\"\"\n",
   1831     "    full_idx = factor_idx.union(prices_idx)\n",
   1832     "\n",
   1833     "    # drop days of the week that are not used\n",
   1834     "    days_to_keep = []\n",
   1835     "    days_of_the_week = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']\n",
   1836     "    for day, day_str in enumerate(days_of_the_week):\n",
   1837     "        if (full_idx.dayofweek == day).any():\n",
   1838     "            days_to_keep.append(day_str)\n",
   1839     "\n",
   1840     "    days_to_keep = ' '.join(days_to_keep)\n",
   1841     "\n",
   1842     "    # we currently don't infer holidays, but CustomBusinessDay class supports\n",
   1843     "    # custom holidays. So holidays could be inferred too eventually\n",
   1844     "    return CustomBusinessDay(weekmask=days_to_keep)\n",
   1845     "\n",
   1846     "def diff_custom_calendar_timedeltas(start, end, freq):\n",
   1847     "    \"\"\"\n",
   1848     "    Compute the difference between two pd.Timedelta taking into consideration\n",
   1849     "    custom frequency, which is used to deal with custom calendars, such as a\n",
   1850     "    trading calendar\n",
   1851     "    Parameters\n",
   1852     "    ----------\n",
   1853     "    start : pd.Timestamp\n",
   1854     "    end : pd.Timestamp\n",
   1855     "    freq : DateOffset, optional\n",
   1856     "    Returns\n",
   1857     "    -------\n",
   1858     "    pd.Timedelta\n",
   1859     "        end - start\n",
   1860     "    \"\"\"\n",
   1861     "    actual_days = pd.date_range(start, end, freq=freq).shape[0] - 1\n",
   1862     "    timediff = end - start\n",
   1863     "    delta_days = timediff.components.days - actual_days\n",
   1864     "    return timediff - pd.Timedelta(days=delta_days)\n",
   1865     "\n",
   1866     "def get_clean_factor(factor,\n",
   1867     "                     forward_returns,\n",
   1868     "                     groupby=None,\n",
   1869     "                     binning_by_group=False,\n",
   1870     "                     quantiles=5,\n",
   1871     "                     bins=None,\n",
   1872     "                     groupby_labels=None,\n",
   1873     "                     max_loss=0.35):\n",
   1874     "    \"\"\"\n",
   1875     "    Formats the factor data, forward return data, and group mappings into a\n",
   1876     "    DataFrame that contains aligned MultiIndex indices of timestamp and asset.\n",
   1877     "    The returned data will be formatted to be suitable for Alphalens functions.\n",
   1878     "    It is safe to skip a call to this function and still make use of Alphalens\n",
   1879     "    functionalities as long as the factor data conforms to the format returned\n",
   1880     "    from get_clean_factor_and_forward_returns and documented here\n",
   1881     "    Parameters\n",
   1882     "    ----------\n",
   1883     "    factor : pd.Series - MultiIndex\n",
   1884     "        A MultiIndex Series indexed by timestamp (level 0) and asset\n",
   1885     "        (level 1), containing the values for a single alpha factor.\n",
   1886     "        ::\n",
   1887     "            -----------------------------------\n",
   1888     "                date    |    asset   |\n",
   1889     "            -----------------------------------\n",
   1890     "                        |   AAPL     |   0.5\n",
   1891     "                        -----------------------\n",
   1892     "                        |   BA       |  -1.1\n",
   1893     "                        -----------------------\n",
   1894     "            2014-01-01  |   CMG      |   1.7\n",
   1895     "                        -----------------------\n",
   1896     "                        |   DAL      |  -0.1\n",
   1897     "                        -----------------------\n",
   1898     "                        |   LULU     |   2.7\n",
   1899     "                        -----------------------\n",
   1900     "    forward_returns : pd.DataFrame - MultiIndex\n",
   1901     "        A MultiIndex DataFrame indexed by timestamp (level 0) and asset\n",
   1902     "        (level 1), containing the forward returns for assets.\n",
   1903     "        Forward returns column names must follow the format accepted by\n",
   1904     "        pd.Timedelta (e.g. '1D', '30m', '3h15m', '1D1h', etc).\n",
   1905     "        'date' index freq property must be set to a trading calendar\n",
   1906     "        (pandas DateOffset), see infer_trading_calendar for more details.\n",
   1907     "        This information is currently used only in cumulative returns\n",
   1908     "        computation\n",
   1909     "        ::\n",
   1910     "            ---------------------------------------\n",
   1911     "                       |       | 1D  | 5D  | 10D\n",
   1912     "            ---------------------------------------\n",
   1913     "                date   | asset |     |     |\n",
   1914     "            ---------------------------------------\n",
   1915     "                       | AAPL  | 0.09|-0.01|-0.079\n",
   1916     "                       ----------------------------\n",
   1917     "                       | BA    | 0.02| 0.06| 0.020\n",
   1918     "                       ----------------------------\n",
   1919     "            2014-01-01 | CMG   | 0.03| 0.09| 0.036\n",
   1920     "                       ----------------------------\n",
   1921     "                       | DAL   |-0.02|-0.06|-0.029\n",
   1922     "                       ----------------------------\n",
   1923     "                       | LULU  |-0.03| 0.05|-0.009\n",
   1924     "                       ----------------------------\n",
   1925     "    groupby : pd.Series - MultiIndex or dict\n",
   1926     "        Either A MultiIndex Series indexed by date and asset,\n",
   1927     "        containing the period wise group codes for each asset, or\n",
   1928     "        a dict of asset to group mappings. If a dict is passed,\n",
   1929     "        it is assumed that group mappings are unchanged for the\n",
   1930     "        entire time period of the passed factor data.\n",
   1931     "    binning_by_group : bool\n",
   1932     "        If True, compute quantile buckets separately for each group.\n",
   1933     "        This is useful when the factor values range vary considerably\n",
   1934     "        across gorups so that it is wise to make the binning group relative.\n",
   1935     "        You should probably enable this if the factor is intended\n",
   1936     "        to be analyzed for a group neutral portfolio\n",
   1937     "    quantiles : int or sequence[float]\n",
   1938     "        Number of equal-sized quantile buckets to use in factor bucketing.\n",
   1939     "        Alternately sequence of quantiles, allowing non-equal-sized buckets\n",
   1940     "        e.g. [0, .10, .5, .90, 1.] or [.05, .5, .95]\n",
   1941     "        Only one of 'quantiles' or 'bins' can be not-None\n",
   1942     "    bins : int or sequence[float]\n",
   1943     "        Number of equal-width (valuewise) bins to use in factor bucketing.\n",
   1944     "        Alternately sequence of bin edges allowing for non-uniform bin width\n",
   1945     "        e.g. [-4, -2, -0.5, 0, 10]\n",
   1946     "        Chooses the buckets to be evenly spaced according to the values\n",
   1947     "        themselves. Useful when the factor contains discrete values.\n",
   1948     "        Only one of 'quantiles' or 'bins' can be not-None\n",
   1949     "    groupby_labels : dict\n",
   1950     "        A dictionary keyed by group code with values corresponding\n",
   1951     "        to the display name for each group.\n",
   1952     "    max_loss : float, optional\n",
   1953     "        Maximum percentage (0.00 to 1.00) of factor data dropping allowed,\n",
   1954     "        computed comparing the number of items in the input factor index and\n",
   1955     "        the number of items in the output DataFrame index.\n",
   1956     "        Factor data can be partially dropped due to being flawed itself\n",
   1957     "        (e.g. NaNs), not having provided enough price data to compute\n",
   1958     "        forward returns for all factor values, or because it is not possible\n",
   1959     "        to perform binning.\n",
   1960     "        Set max_loss=0 to avoid Exceptions suppression.\n",
   1961     "    Returns\n",
   1962     "    -------\n",
   1963     "    merged_data : pd.DataFrame - MultiIndex\n",
   1964     "        A MultiIndex Series indexed by date (level 0) and asset (level 1),\n",
   1965     "        containing the values for a single alpha factor, forward returns for\n",
   1966     "        each period, the factor quantile/bin that factor value belongs to, and\n",
   1967     "        (optionally) the group the asset belongs to.\n",
   1968     "        - forward returns column names follow the format accepted by\n",
   1969     "          pd.Timedelta (e.g. '1D', '30m', '3h15m', '1D1h', etc)\n",
   1970     "        - 'date' index freq property (merged_data.index.levels[0].freq) is the\n",
   1971     "          same as that of the input forward returns data. This is currently\n",
   1972     "          used only in cumulative returns computation\n",
   1973     "        ::\n",
   1974     "           -------------------------------------------------------------------\n",
   1975     "                      |       | 1D  | 5D  | 10D  |factor|group|factor_quantile\n",
   1976     "           -------------------------------------------------------------------\n",
   1977     "               date   | asset |     |     |      |      |     |\n",
   1978     "           -------------------------------------------------------------------\n",
   1979     "                      | AAPL  | 0.09|-0.01|-0.079|  0.5 |  G1 |      3\n",
   1980     "                      --------------------------------------------------------\n",
   1981     "                      | BA    | 0.02| 0.06| 0.020| -1.1 |  G2 |      5\n",
   1982     "                      --------------------------------------------------------\n",
   1983     "           2014-01-01 | CMG   | 0.03| 0.09| 0.036|  1.7 |  G2 |      1\n",
   1984     "                      --------------------------------------------------------\n",
   1985     "                      | DAL   |-0.02|-0.06|-0.029| -0.1 |  G3 |      5\n",
   1986     "                      --------------------------------------------------------\n",
   1987     "                      | LULU  |-0.03| 0.05|-0.009|  2.7 |  G1 |      2\n",
   1988     "                      --------------------------------------------------------\n",
   1989     "    \"\"\"\n",
   1990     "\n",
   1991     "    initial_amount = float(len(factor.index))\n",
   1992     "\n",
   1993     "    factor = factor.copy()\n",
   1994     "    factor.index = factor.index.rename(['date', 'asset'])\n",
   1995     "\n",
   1996     "    merged_data = forward_returns.copy()\n",
   1997     "    merged_data['factor'] = factor\n",
   1998     "\n",
   1999     "    if groupby is not None:\n",
   2000     "        if isinstance(groupby, dict):\n",
   2001     "            diff = set(factor.index.get_level_values(\n",
   2002     "                'asset')) - set(groupby.keys())\n",
   2003     "            if len(diff) > 0:\n",
   2004     "                raise KeyError(\n",
   2005     "                    \"Assets {} not in group mapping\".format(\n",
   2006     "                        list(diff)))\n",
   2007     "\n",
   2008     "            ss = pd.Series(groupby)\n",
   2009     "            groupby = pd.Series(index=factor.index,\n",
   2010     "                                data=ss[factor.index.get_level_values(\n",
   2011     "                                    'asset')].values)\n",
   2012     "\n",
   2013     "        if groupby_labels is not None:\n",
   2014     "            diff = set(groupby.values) - set(groupby_labels.keys())\n",
   2015     "            if len(diff) > 0:\n",
   2016     "                raise KeyError(\n",
   2017     "                    \"groups {} not in passed group names\".format(\n",
   2018     "                        list(diff)))\n",
   2019     "\n",
   2020     "            sn = pd.Series(groupby_labels)\n",
   2021     "            groupby = pd.Series(index=groupby.index,\n",
   2022     "                                data=sn[groupby.values].values)\n",
   2023     "\n",
   2024     "        merged_data['group'] = groupby.astype('category')\n",
   2025     "\n",
   2026     "    merged_data = merged_data.dropna()\n",
   2027     "\n",
   2028     "    fwdret_amount = float(len(merged_data.index))\n",
   2029     "\n",
   2030     "    no_raise = False if max_loss == 0 else True\n",
   2031     "    merged_data['factor_quantile'] = quantize_factor(merged_data,\n",
   2032     "                                                     quantiles,\n",
   2033     "                                                     bins,\n",
   2034     "                                                     binning_by_group,\n",
   2035     "                                                     no_raise)\n",
   2036     "\n",
   2037     "    merged_data = merged_data.dropna()\n",
   2038     "\n",
   2039     "    binning_amount = float(len(merged_data.index))\n",
   2040     "\n",
   2041     "    tot_loss = (initial_amount - binning_amount) / initial_amount\n",
   2042     "    fwdret_loss = (initial_amount - fwdret_amount) / initial_amount\n",
   2043     "    bin_loss = tot_loss - fwdret_loss\n",
   2044     "\n",
   2045     "#     print(\"Dropped %.1f%% entries from factor data: %.1f%% in forward \"\n",
   2046     "#           \"returns computation and %.1f%% in binning phase \"\n",
   2047     "#           \"(set max_loss=0 to see potentially suppressed Exceptions).\" %\n",
   2048     "#           (tot_loss * 100, fwdret_loss * 100,  bin_loss * 100))\n",
   2049     "\n",
   2050     "    if tot_loss > max_loss:\n",
   2051     "        message = (\"max_loss (%.1f%%) exceeded %.1f%%, consider increasing it.\"\n",
   2052     "                   % (max_loss * 100, tot_loss * 100))\n",
   2053     "        raise MaxLossExceededError(message)\n",
   2054     "    else:\n",
   2055     "#         print(\"max_loss is %.1f%%, not exceeded: OK!\" % (max_loss * 100))\n",
   2056     "        pass\n",
   2057     "\n",
   2058     "    return merged_data\n",
   2059     "\n",
   2060     "def timedelta_to_string(timedelta):\n",
   2061     "    \"\"\"\n",
   2062     "    Utility that converts a pandas.Timedelta to a string representation\n",
   2063     "    compatible with pandas.Timedelta constructor format\n",
   2064     "    Parameters\n",
   2065     "    ----------\n",
   2066     "    timedelta: pd.Timedelta\n",
   2067     "    Returns\n",
   2068     "    -------\n",
   2069     "    string\n",
   2070     "        string representation of 'timedelta'\n",
   2071     "    \"\"\"\n",
   2072     "    c = timedelta.components\n",
   2073     "    format = ''\n",
   2074     "    if c.days != 0:\n",
   2075     "        format += '%dD' % c.days\n",
   2076     "    if c.hours > 0:\n",
   2077     "        format += '%dh' % c.hours\n",
   2078     "    if c.minutes > 0:\n",
   2079     "        format += '%dm' % c.minutes\n",
   2080     "    if c.seconds > 0:\n",
   2081     "        format += '%ds' % c.seconds\n",
   2082     "    if c.milliseconds > 0:\n",
   2083     "        format += '%dms' % c.milliseconds\n",
   2084     "    if c.microseconds > 0:\n",
   2085     "        format += '%dus' % c.microseconds\n",
   2086     "    if c.nanoseconds > 0:\n",
   2087     "        format += '%dns' % c.nanoseconds\n",
   2088     "    return format\n",
   2089     "\n",
   2090     "\n",
   2091     "def add_custom_calendar_timedelta(inputs, timedelta, freq):\n",
   2092     "    \"\"\"\n",
   2093     "    Add timedelta to 'input' taking into consideration custom frequency, which\n",
   2094     "    is used to deal with custom calendars, such as a trading calendar\n",
   2095     "    Parameters\n",
   2096     "    ----------\n",
   2097     "    input : pd.DatetimeIndex or pd.Timestamp\n",
   2098     "    timedelta : pd.Timedelta\n",
   2099     "    freq : DateOffset, optional\n",
   2100     "    Returns\n",
   2101     "    -------\n",
   2102     "    pd.DatetimeIndex or pd.Timestamp\n",
   2103     "        input + timedelta\n",
   2104     "    \"\"\"\n",
   2105     "    days = timedelta.components.days\n",
   2106     "    offset = timedelta - pd.Timedelta(days=days)\n",
   2107     "    return inputs + freq * days + offset\n",
   2108     "\n",
   2109     "def non_unique_bin_edges_error(func):\n",
   2110     "    \"\"\"\n",
   2111     "    Give user a more informative error in case it is not possible\n",
   2112     "    to properly calculate quantiles on the input dataframe (factor)\n",
   2113     "    \"\"\"\n",
   2114     "    message = \"\"\"\n",
   2115     "    An error occurred while computing bins/quantiles on the input provided.\n",
   2116     "    This usually happens when the input contains too many identical\n",
   2117     "    values and they span more than one quantile. The quantiles are choosen\n",
   2118     "    to have the same number of records each, but the same value cannot span\n",
   2119     "    multiple quantiles. Possible workarounds are:\n",
   2120     "    1 - Decrease the number of quantiles\n",
   2121     "    2 - Specify a custom quantiles range, e.g. [0, .50, .75, 1.] to get unequal\n",
   2122     "        number of records per quantile\n",
   2123     "    3 - Use 'bins' option instead of 'quantiles', 'bins' chooses the\n",
   2124     "        buckets to be evenly spaced according to the values themselves, while\n",
   2125     "        'quantiles' forces the buckets to have the same number of records.\n",
   2126     "    4 - for factors with discrete values use the 'bins' option with custom\n",
   2127     "        ranges and create a range for each discrete value\n",
   2128     "    Please see utils.get_clean_factor_and_forward_returns documentation for\n",
   2129     "    full documentation of 'bins' and 'quantiles' options.\n",
   2130     "\"\"\"\n",
   2131     "\n",
   2132     "    def dec(*args, **kwargs):\n",
   2133     "        try:\n",
   2134     "            return func(*args, **kwargs)\n",
   2135     "        except ValueError as e:\n",
   2136     "            if 'Bin edges must be unique' in str(e):\n",
   2137     "                rethrow(e, message)\n",
   2138     "            raise\n",
   2139     "    return dec\n",
   2140     "\n",
   2141     "@non_unique_bin_edges_error\n",
   2142     "def quantize_factor(factor_data,\n",
   2143     "                    quantiles=5,\n",
   2144     "                    bins=None,\n",
   2145     "                    by_group=False,\n",
   2146     "                    no_raise=False):\n",
   2147     "    \"\"\"\n",
   2148     "    Computes period wise factor quantiles.\n",
   2149     "    Parameters\n",
   2150     "    ----------\n",
   2151     "    factor_data : pd.DataFrame - MultiIndex\n",
   2152     "        A MultiIndex DataFrame indexed by date (level 0) and asset (level 1),\n",
   2153     "        containing the values for a single alpha factor, forward returns for\n",
   2154     "        each period, the factor quantile/bin that factor value belongs to, and\n",
   2155     "        (optionally) the group the asset belongs to.\n",
   2156     "        - See full explanation in utils.get_clean_factor_and_forward_returns\n",
   2157     "    quantiles : int or sequence[float]\n",
   2158     "        Number of equal-sized quantile buckets to use in factor bucketing.\n",
   2159     "        Alternately sequence of quantiles, allowing non-equal-sized buckets\n",
   2160     "        e.g. [0, .10, .5, .90, 1.] or [.05, .5, .95]\n",
   2161     "        Only one of 'quantiles' or 'bins' can be not-None\n",
   2162     "    bins : int or sequence[float]\n",
   2163     "        Number of equal-width (valuewise) bins to use in factor bucketing.\n",
   2164     "        Alternately sequence of bin edges allowing for non-uniform bin width\n",
   2165     "        e.g. [-4, -2, -0.5, 0, 10]\n",
   2166     "        Only one of 'quantiles' or 'bins' can be not-None\n",
   2167     "    by_group : bool\n",
   2168     "        If True, compute quantile buckets separately for each group.\n",
   2169     "    no_raise: bool, optional\n",
   2170     "        If True, no exceptions are thrown and the values for which the\n",
   2171     "        exception would have been thrown are set to np.NaN\n",
   2172     "    Returns\n",
   2173     "    -------\n",
   2174     "    factor_quantile : pd.Series\n",
   2175     "        Factor quantiles indexed by date and asset.\n",
   2176     "    \"\"\"\n",
   2177     "    if not ((quantiles is not None and bins is None) or\n",
   2178     "            (quantiles is None and bins is not None)):\n",
   2179     "        raise ValueError('Either quantiles or bins should be provided')\n",
   2180     "\n",
   2181     "    def quantile_calc(x, _quantiles, _bins, _no_raise):\n",
   2182     "        try:\n",
   2183     "            if _quantiles is not None and _bins is None:\n",
   2184     "                return pd.qcut(x, _quantiles, labels=False) + 1\n",
   2185     "            elif _bins is not None and _quantiles is None:\n",
   2186     "                return pd.cut(x, _bins, labels=False) + 1\n",
   2187     "        except Exception as e:\n",
   2188     "            if _no_raise:\n",
   2189     "                return pd.Series(index=x.index)\n",
   2190     "            raise e\n",
   2191     "\n",
   2192     "    grouper = [factor_data.index.get_level_values('date')]\n",
   2193     "    if by_group:\n",
   2194     "        grouper.append('group')\n",
   2195     "\n",
   2196     "    factor_quantile = factor_data.groupby(grouper)['factor'] \\\n",
   2197     "        .apply(quantile_calc, quantiles, bins, no_raise)\n",
   2198     "    factor_quantile.name = 'factor_quantile'\n",
   2199     "\n",
   2200     "    return factor_quantile.dropna()"
   2201    ]
   2202   },
   2203   {
   2204    "cell_type": "code",
   2205    "execution_count": 16,
   2206    "metadata": {},
   2207    "outputs": [],
   2208    "source": [
   2209     "def new_is_positive():\n",
   2210     "    return False"
   2211    ]
   2212   },
   2213   {
   2214    "cell_type": "code",
   2215    "execution_count": null,
   2216    "metadata": {
   2217     "collapsed": true
   2218    },
   2219    "outputs": [],
   2220    "source": []
   2221   }
   2222  ],
   2223  "metadata": {
   2224   "kernelspec": {
   2225    "display_name": "Python 2",
   2226    "language": "python",
   2227    "name": "python2"
   2228   },
   2229   "language_info": {
   2230    "codemirror_mode": {
   2231     "name": "ipython",
   2232     "version": 2
   2233    },
   2234    "file_extension": ".py",
   2235    "mimetype": "text/x-python",
   2236    "name": "python",
   2237    "nbconvert_exporter": "python",
   2238    "pygments_lexer": "ipython2",
   2239    "version": "2.7.12"
   2240   }
   2241  },
   2242  "nbformat": 4,
   2243  "nbformat_minor": 2
   2244 }