ml-finance-python
python scripts for finance machine learning
git clone https://9o.is/git/ml-finance-python.git
04_logistic_regression.ipynb
(388623B)
1 {
2 "cells": [
3 {
4 "cell_type": "markdown",
5 "metadata": {},
6 "source": [
7 "# How to use logistic regression for prediction"
8 ]
9 },
10 {
11 "cell_type": "markdown",
12 "metadata": {},
13 "source": [
14 "The lasso L1 penalty and the ridge L2 penalty can both be used with logistic regression. They have the same shrinkage effect as we have just discussed, and the lasso can again be used for variable selection with any linear regression model.\n",
15 "\n",
16 "Just as with linear regression, it is important to standardize the input variables as the regularized models are scale sensitive. The regularization hyperparameter also requires tuning using cross-validation as in the linear regression case."
17 ]
18 },
19 {
20 "cell_type": "markdown",
21 "metadata": {},
22 "source": [
23 "## How to predict price movements using sklearn"
24 ]
25 },
26 {
27 "cell_type": "markdown",
28 "metadata": {},
29 "source": [
30 "We continue the price prediction example but now we binarize the outcome variable so that it takes on the value 1 whenever the 10-day return is positive and 0 otherwise; see the notebook logistic_regression.ipynb in the sub directory stock_price_prediction"
31 ]
32 },
33 {
34 "cell_type": "markdown",
35 "metadata": {},
36 "source": [
37 "## Imports"
38 ]
39 },
40 {
41 "cell_type": "code",
42 "execution_count": 66,
43 "metadata": {},
44 "outputs": [],
45 "source": [
46 "import pandas as pd\n",
47 "import numpy as np\n",
48 "from time import time\n",
49 "import talib\n",
50 "import re\n",
51 "from statsmodels.api import OLS\n",
52 "from sklearn.metrics import mean_squared_error\n",
53 "from scipy.stats import spearmanr\n",
54 "from sklearn.linear_model import LinearRegression, Ridge, RidgeCV, Lasso, LassoCV, LogisticRegression\n",
55 "from sklearn.preprocessing import StandardScaler\n",
56 "\n",
57 "from quantopian.research import run_pipeline\n",
58 "from quantopian.pipeline import Pipeline, factors, filters, classifiers\n",
59 "from quantopian.pipeline.data.builtin import USEquityPricing\n",
60 "from quantopian.pipeline.factors import (Latest, \n",
61 " Returns, \n",
62 " AverageDollarVolume, \n",
63 " SimpleMovingAverage,\n",
64 " EWMA,\n",
65 " BollingerBands,\n",
66 " CustomFactor,\n",
67 " MarketCap,\n",
68 " SimpleBeta)\n",
69 "from quantopian.pipeline.filters import QTradableStocksUS, StaticAssets\n",
70 "from quantopian.pipeline.data.quandl import fred_usdontd156n as libor\n",
71 "from empyrical import max_drawdown, sortino_ratio\n",
72 "\n",
73 "import seaborn as sns\n",
74 "import matplotlib.pyplot as plt\n",
75 "from sklearn.metrics import roc_auc_score"
76 ]
77 },
78 {
79 "cell_type": "markdown",
80 "metadata": {},
81 "source": [
82 "## Data Sources"
83 ]
84 },
85 {
86 "cell_type": "code",
87 "execution_count": 2,
88 "metadata": {},
89 "outputs": [],
90 "source": [
91 "################\n",
92 "# Fundamentals #\n",
93 "################\n",
94 "\n",
95 "# Morningstar fundamentals (2002 - Ongoing)\n",
96 "# https://www.quantopian.com/help/fundamentals\n",
97 "from quantopian.pipeline.data import Fundamentals\n",
98 "\n",
99 "#####################\n",
100 "# Analyst Estimates #\n",
101 "#####################\n",
102 "\n",
103 "# Earnings Surprises - Zacks (27 May 2006 - Ongoing)\n",
104 "# https://www.quantopian.com/data/zacks/earnings_surprises\n",
105 "from quantopian.pipeline.data.zacks import EarningsSurprises\n",
106 "from quantopian.pipeline.factors.zacks import BusinessDaysSinceEarningsSurprisesAnnouncement\n",
107 "\n",
108 "##########\n",
109 "# Events #\n",
110 "##########\n",
111 "\n",
112 "# Buyback Announcements - EventVestor (01 Jun 2007 - Ongoing)\n",
113 "# https://www.quantopian.com/data/eventvestor/buyback_auth\n",
114 "from quantopian.pipeline.data.eventvestor import BuybackAuthorizations\n",
115 "from quantopian.pipeline.factors.eventvestor import BusinessDaysSinceBuybackAuth\n",
116 "\n",
117 "# CEO Changes - EventVestor (01 Jan 2007 - Ongoing)\n",
118 "# https://www.quantopian.com/data/eventvestor/ceo_change\n",
119 "from quantopian.pipeline.data.eventvestor import CEOChangeAnnouncements\n",
120 "\n",
121 "# Dividends - EventVestor (01 Jan 2007 - Ongoing)\n",
122 "# https://www.quantopian.com/data/eventvestor/dividends\n",
123 "from quantopian.pipeline.data.eventvestor import (\n",
124 " DividendsByExDate,\n",
125 " DividendsByPayDate,\n",
126 " DividendsByAnnouncementDate,\n",
127 ")\n",
128 "from quantopian.pipeline.factors.eventvestor import (\n",
129 " BusinessDaysSincePreviousExDate,\n",
130 " BusinessDaysUntilNextExDate,\n",
131 " BusinessDaysSinceDividendAnnouncement,\n",
132 ")\n",
133 "\n",
134 "# Earnings Calendar - EventVestor (01 Jan 2007 - Ongoing)\n",
135 "# https://www.quantopian.com/data/eventvestor/earnings_calendar\n",
136 "from quantopian.pipeline.data.eventvestor import EarningsCalendar\n",
137 "from quantopian.pipeline.factors.eventvestor import (\n",
138 " BusinessDaysUntilNextEarnings,\n",
139 " BusinessDaysSincePreviousEarnings\n",
140 ")\n",
141 "\n",
142 "# 13D Filings - EventVestor (01 Jan 2007 - Ongoing)\n",
143 "# https://www.quantopian.com/data/eventvestor/_13d_filings\n",
144 "from quantopian.pipeline.data.eventvestor import _13DFilings\n",
145 "from quantopian.pipeline.factors.eventvestor import BusinessDaysSince13DFilingsDate\n",
146 "\n",
147 "#############\n",
148 "# Sentiment #\n",
149 "#############\n",
150 "\n",
151 "# News Sentiment - Sentdex Sentiment Analysis (15 Oct 2012 - Ongoing)\n",
152 "# https://www.quantopian.com/data/sentdex/sentiment\n",
153 "from quantopian.pipeline.data.sentdex import sentiment"
154 ]
155 },
156 {
157 "cell_type": "markdown",
158 "metadata": {},
159 "source": [
160 "## Setup"
161 ]
162 },
163 {
164 "cell_type": "markdown",
165 "metadata": {},
166 "source": [
167 "### Time Horizon"
168 ]
169 },
170 {
171 "cell_type": "code",
172 "execution_count": 3,
173 "metadata": {},
174 "outputs": [],
175 "source": [
176 "# trading days per period\n",
177 "MONTH = 21\n",
178 "YEAR = 12 * MONTH"
179 ]
180 },
181 {
182 "cell_type": "code",
183 "execution_count": 4,
184 "metadata": {},
185 "outputs": [],
186 "source": [
187 "START = '2014-01-01'\n",
188 "END = '2015-12-31'"
189 ]
190 },
191 {
192 "cell_type": "markdown",
193 "metadata": {},
194 "source": [
195 "### Universe"
196 ]
197 },
198 {
199 "cell_type": "code",
200 "execution_count": 5,
201 "metadata": {},
202 "outputs": [],
203 "source": [
204 "def Q100US():\n",
205 " return filters.make_us_equity_universe(\n",
206 " target_size=100,\n",
207 " rankby=factors.AverageDollarVolume(window_length=200),\n",
208 " mask=filters.default_us_equity_universe_mask(),\n",
209 " groupby=classifiers.fundamentals.Sector(),\n",
210 " max_group_weight=0.3,\n",
211 " smoothing_func=lambda f: f.downsample('month_start'),\n",
212 " )"
213 ]
214 },
215 {
216 "cell_type": "code",
217 "execution_count": 6,
218 "metadata": {},
219 "outputs": [],
220 "source": [
221 "# UNIVERSE = StaticAssets(symbols(['MSFT', 'AAPL']))\n",
222 "UNIVERSE = Q100US()"
223 ]
224 },
225 {
226 "cell_type": "code",
227 "execution_count": 7,
228 "metadata": {},
229 "outputs": [],
230 "source": [
231 "class AnnualizedData(CustomFactor):\n",
232 " # Get the sum of the last 4 reported values\n",
233 " window_length = 260\n",
234 "\n",
235 " def compute(self, today, assets, out, asof_date, values):\n",
236 " for asset in range(len(assets)):\n",
237 " # unique asof dates indicate availability of new figures\n",
238 " _, filing_dates = np.unique(asof_date[:, asset], return_index=True)\n",
239 " quarterly_values = values[filing_dates[-4:], asset]\n",
240 " # ignore annual windows with <4 quarterly data points\n",
241 " if len(~np.isnan(quarterly_values)) != 4:\n",
242 " out[asset] = np.nan\n",
243 " else:\n",
244 " out[asset] = np.sum(quarterly_values)"
245 ]
246 },
247 {
248 "cell_type": "code",
249 "execution_count": 8,
250 "metadata": {},
251 "outputs": [],
252 "source": [
253 "class AnnualAvg(CustomFactor):\n",
254 " window_length = 252\n",
255 " \n",
256 " def compute(self, today, assets, out, values):\n",
257 " out[:] = (values[0] + values[-1])/2"
258 ]
259 },
260 {
261 "cell_type": "code",
262 "execution_count": 9,
263 "metadata": {},
264 "outputs": [],
265 "source": [
266 "def factor_pipeline(factors):\n",
267 " start = time()\n",
268 " pipe = Pipeline({k: v(mask=UNIVERSE).rank() for k, v in factors.items()},\n",
269 " screen=UNIVERSE)\n",
270 " result = run_pipeline(pipe, start_date=START, end_date=END)\n",
271 " return result, time() - start"
272 ]
273 },
274 {
275 "cell_type": "markdown",
276 "metadata": {},
277 "source": [
278 "## Factor Library"
279 ]
280 },
281 {
282 "cell_type": "markdown",
283 "metadata": {},
284 "source": [
285 "### Value Factors"
286 ]
287 },
288 {
289 "cell_type": "code",
290 "execution_count": 10,
291 "metadata": {},
292 "outputs": [],
293 "source": [
294 "class ValueFactors:\n",
295 " \"\"\"Definitions of factors for cross-sectional trading algorithms\"\"\"\n",
296 " \n",
297 " @staticmethod\n",
298 " def PriceToSalesTTM(**kwargs):\n",
299 " \"\"\"Last closing price divided by sales per share\"\"\" \n",
300 " return Fundamentals.ps_ratio.latest\n",
301 "\n",
302 " @staticmethod\n",
303 " def PriceToEarningsTTM(**kwargs):\n",
304 " \"\"\"Closing price divided by earnings per share (EPS)\"\"\"\n",
305 " return Fundamentals.pe_ratio.latest\n",
306 " \n",
307 " @staticmethod\n",
308 " def PriceToDilutedEarningsTTM(mask):\n",
309 " \"\"\"Closing price divided by diluted EPS\"\"\"\n",
310 " last_close = USEquityPricing.close.latest\n",
311 " diluted_eps = AnnualizedData(inputs = [Fundamentals.diluted_eps_earnings_reports_asof_date,\n",
312 " Fundamentals.diluted_eps_earnings_reports],\n",
313 " mask=mask)\n",
314 " return last_close / diluted_eps\n",
315 "\n",
316 " @staticmethod\n",
317 " def PriceToForwardEarnings(**kwargs):\n",
318 " \"\"\"Price to Forward Earnings\"\"\"\n",
319 " return Fundamentals.forward_pe_ratio.latest\n",
320 " \n",
321 " @staticmethod\n",
322 " def DividendYield(**kwargs):\n",
323 " \"\"\"Dividends per share divided by closing price\"\"\"\n",
324 " return Fundamentals.trailing_dividend_yield.latest\n",
325 "\n",
326 " @staticmethod\n",
327 " def PriceToFCF(mask):\n",
328 " \"\"\"Price to Free Cash Flow\"\"\"\n",
329 " last_close = USEquityPricing.close.latest\n",
330 " fcf_share = AnnualizedData(inputs = [Fundamentals.fcf_per_share_asof_date,\n",
331 " Fundamentals.fcf_per_share],\n",
332 " mask=mask)\n",
333 " return last_close / fcf_share\n",
334 "\n",
335 " @staticmethod\n",
336 " def PriceToOperatingCashflow(mask):\n",
337 " \"\"\"Last Close divided by Operating Cash Flows\"\"\"\n",
338 " last_close = USEquityPricing.close.latest\n",
339 " cfo_per_share = AnnualizedData(inputs = [Fundamentals.cfo_per_share_asof_date,\n",
340 " Fundamentals.cfo_per_share],\n",
341 " mask=mask) \n",
342 " return last_close / cfo_per_share\n",
343 "\n",
344 " @staticmethod\n",
345 " def PriceToBook(mask):\n",
346 " \"\"\"Closing price divided by book value\"\"\"\n",
347 " last_close = USEquityPricing.close.latest\n",
348 " book_value_per_share = AnnualizedData(inputs = [Fundamentals.book_value_per_share_asof_date,\n",
349 " Fundamentals.book_value_per_share],\n",
350 " mask=mask) \n",
351 " return last_close / book_value_per_share\n",
352 "\n",
353 "\n",
354 " @staticmethod\n",
355 " def EVToFCF(mask):\n",
356 " \"\"\"Enterprise Value divided by Free Cash Flows\"\"\"\n",
357 " fcf = AnnualizedData(inputs = [Fundamentals.free_cash_flow_asof_date,\n",
358 " Fundamentals.free_cash_flow],\n",
359 " mask=mask)\n",
360 " return Fundamentals.enterprise_value.latest / fcf\n",
361 "\n",
362 " @staticmethod\n",
363 " def EVToEBITDA(mask):\n",
364 " \"\"\"Enterprise Value to Earnings Before Interest, Taxes, Deprecation and Amortization (EBITDA)\"\"\"\n",
365 " ebitda = AnnualizedData(inputs = [Fundamentals.ebitda_asof_date,\n",
366 " Fundamentals.ebitda],\n",
367 " mask=mask)\n",
368 "\n",
369 " return Fundamentals.enterprise_value.latest / ebitda\n",
370 "\n",
371 " @staticmethod\n",
372 " def EBITDAYield(mask):\n",
373 " \"\"\"EBITDA divided by latest close\"\"\"\n",
374 " ebitda = AnnualizedData(inputs = [Fundamentals.ebitda_asof_date,\n",
375 " Fundamentals.ebitda],\n",
376 " mask=mask)\n",
377 " return USEquityPricing.close.latest / ebitda"
378 ]
379 },
380 {
381 "cell_type": "code",
382 "execution_count": 11,
383 "metadata": {},
384 "outputs": [],
385 "source": [
386 "VALUE_FACTORS = {\n",
387 " 'DividendYield' : ValueFactors.DividendYield,\n",
388 " 'EBITDAYield' : ValueFactors.EBITDAYield,\n",
389 " 'EVToEBITDA' : ValueFactors.EVToEBITDA,\n",
390 " 'EVToFCF' : ValueFactors.EVToFCF,\n",
391 " 'PriceToBook' : ValueFactors.PriceToBook,\n",
392 " 'PriceToDilutedEarningsTTM': ValueFactors.PriceToDilutedEarningsTTM,\n",
393 " 'PriceToEarningsTTM' : ValueFactors.PriceToEarningsTTM,\n",
394 " 'PriceToFCF' : ValueFactors.PriceToFCF,\n",
395 " 'PriceToForwardEarnings' : ValueFactors.PriceToForwardEarnings,\n",
396 " 'PriceToOperatingCashflow' : ValueFactors.PriceToOperatingCashflow,\n",
397 " 'PriceToSalesTTM' : ValueFactors.PriceToSalesTTM,\n",
398 "}"
399 ]
400 },
401 {
402 "cell_type": "code",
403 "execution_count": 12,
404 "metadata": {},
405 "outputs": [
406 {
407 "name": "stderr",
408 "output_type": "stream",
409 "text": [
410 "/usr/local/lib/python2.7/dist-packages/numpy/lib/arraysetops.py:200: FutureWarning: In the future, NAT != NAT will be True rather than False.\n",
411 " flag = np.concatenate(([True], aux[1:] != aux[:-1]))\n"
412 ]
413 },
414 {
415 "name": "stdout",
416 "output_type": "stream",
417 "text": [
418 "Pipeline run time 96.25 secs\n",
419 "<class 'pandas.core.frame.DataFrame'>\n",
420 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
421 "Data columns (total 11 columns):\n",
422 "DividendYield 40772 non-null float64\n",
423 "EBITDAYield 49823 non-null float64\n",
424 "EVToEBITDA 49823 non-null float64\n",
425 "EVToFCF 46400 non-null float64\n",
426 "PriceToBook 50343 non-null float64\n",
427 "PriceToDilutedEarningsTTM 50215 non-null float64\n",
428 "PriceToEarningsTTM 48956 non-null float64\n",
429 "PriceToFCF 49133 non-null float64\n",
430 "PriceToForwardEarnings 39607 non-null float64\n",
431 "PriceToOperatingCashflow 50343 non-null float64\n",
432 "PriceToSalesTTM 50362 non-null float64\n",
433 "dtypes: float64(11)\n",
434 "memory usage: 4.6+ MB\n"
435 ]
436 }
437 ],
438 "source": [
439 "value_result, t = factor_pipeline(VALUE_FACTORS)\n",
440 "print('Pipeline run time {:.2f} secs'.format(t))\n",
441 "value_result.info()"
442 ]
443 },
444 {
445 "cell_type": "markdown",
446 "metadata": {},
447 "source": [
448 "### Momentum"
449 ]
450 },
451 {
452 "cell_type": "code",
453 "execution_count": 13,
454 "metadata": {},
455 "outputs": [],
456 "source": [
457 "class MomentumFactors:\n",
458 " \"\"\"Custom Momentum Factors\"\"\"\n",
459 " class PercentAboveLow(CustomFactor):\n",
460 " \"\"\"Percentage of current close above low \n",
461 " in lookback window of window_length days\n",
462 " \"\"\"\n",
463 " inputs = [USEquityPricing.close]\n",
464 " window_length = 252\n",
465 "\n",
466 " def compute(self, today, assets, out, close):\n",
467 " out[:] = close[-1] / np.min(close, axis=0) - 1\n",
468 "\n",
469 " class PercentBelowHigh(CustomFactor):\n",
470 " \"\"\"Percentage of current close below high \n",
471 " in lookback window of window_length days\n",
472 " \"\"\"\n",
473 " \n",
474 " inputs = [USEquityPricing.close]\n",
475 " window_length = 252\n",
476 " \n",
477 " def compute(self, today, assets, out, close):\n",
478 " out[:] = close[-1] / np.max(close, axis=0) - 1\n",
479 "\n",
480 " @staticmethod\n",
481 " def make_dx(timeperiod=14):\n",
482 " class DX(CustomFactor):\n",
483 " \"\"\"Directional Movement Index\"\"\"\n",
484 " inputs = [USEquityPricing.high, \n",
485 " USEquityPricing.low, \n",
486 " USEquityPricing.close]\n",
487 " window_length = timeperiod + 1\n",
488 " \n",
489 " def compute(self, today, assets, out, high, low, close):\n",
490 " out[:] = [talib.DX(high[:, i], \n",
491 " low[:, i], \n",
492 " close[:, i], \n",
493 " timeperiod=timeperiod)[-1] \n",
494 " for i in range(len(assets))]\n",
495 " return DX \n",
496 "\n",
497 " @staticmethod\n",
498 " def make_mfi(timeperiod=14):\n",
499 " class MFI(CustomFactor):\n",
500 " \"\"\"Money Flow Index\"\"\"\n",
501 " inputs = [USEquityPricing.high, \n",
502 " USEquityPricing.low, \n",
503 " USEquityPricing.close,\n",
504 " USEquityPricing.volume]\n",
505 " window_length = timeperiod + 1\n",
506 " \n",
507 " def compute(self, today, assets, out, high, low, close, vol):\n",
508 " out[:] = [talib.MFI(high[:, i], \n",
509 " low[:, i], \n",
510 " close[:, i],\n",
511 " vol[:, i],\n",
512 " timeperiod=timeperiod)[-1] \n",
513 " for i in range(len(assets))]\n",
514 " return MFI \n",
515 "\n",
516 " @staticmethod\n",
517 " def make_oscillator(fastperiod=12, slowperiod=26, matype=0):\n",
518 " class PPO(CustomFactor):\n",
519 " \"\"\"12/26-Day Percent Price Oscillator\"\"\"\n",
520 " inputs = [USEquityPricing.close]\n",
521 " window_length = slowperiod\n",
522 "\n",
523 " def compute(self, today, assets, out, close_prices):\n",
524 " out[:] = [talib.PPO(close,\n",
525 " fastperiod=fastperiod,\n",
526 " slowperiod=slowperiod, \n",
527 " matype=matype)[-1]\n",
528 " for close in close_prices.T]\n",
529 " return PPO\n",
530 "\n",
531 " @staticmethod\n",
532 " def make_stochastic_oscillator(fastk_period=5, slowk_period=3, slowd_period=3, \n",
533 " slowk_matype=0, slowd_matype=0): \n",
534 " class StochasticOscillator(CustomFactor):\n",
535 " \"\"\"20-day Stochastic Oscillator \"\"\"\n",
536 " inputs = [USEquityPricing.high, \n",
537 " USEquityPricing.low, \n",
538 " USEquityPricing.close]\n",
539 " outputs = ['slowk', 'slowd']\n",
540 " window_length = fastk_period * 2\n",
541 " \n",
542 " def compute(self, today, assets, out, high, low, close):\n",
543 " slowk, slowd = [talib.STOCH(high[:, i],\n",
544 " low[:, i],\n",
545 " close[:, i], \n",
546 " fastk_period=fastk_period,\n",
547 " slowk_period=slowk_period, \n",
548 " slowk_matype=slowk_matype, \n",
549 " slowd_period=slowd_period, \n",
550 " slowd_matype=slowd_matype)[-1] \n",
551 " for i in range(len(assets))]\n",
552 "\n",
553 " out.slowk[:], out.slowd[:] = slowk[-1], slowd[-1]\n",
554 " return StochasticOscillator\n",
555 " \n",
556 " @staticmethod\n",
557 " def make_trendline(timeperiod=252): \n",
558 " class Trendline(CustomFactor):\n",
559 " inputs = [USEquityPricing.close]\n",
560 " \"\"\"52-Week Trendline\"\"\"\n",
561 " window_length = timeperiod\n",
562 "\n",
563 " def compute(self, today, assets, out, close_prices):\n",
564 " out[:] = [talib.LINEARREG_SLOPE(close, \n",
565 " timeperiod=timeperiod)[-1] \n",
566 " for close in close_prices.T]\n",
567 " return Trendline"
568 ]
569 },
570 {
571 "cell_type": "code",
572 "execution_count": 14,
573 "metadata": {},
574 "outputs": [],
575 "source": [
576 "MOMENTUM_FACTORS = {\n",
577 " 'Percent Above Low' : MomentumFactors.PercentAboveLow,\n",
578 " 'Percent Below High' : MomentumFactors.PercentBelowHigh,\n",
579 " 'Price Oscillator' : MomentumFactors.make_oscillator(),\n",
580 " 'Money Flow Index' : MomentumFactors.make_mfi(),\n",
581 " 'Directional Movement Index' : MomentumFactors.make_dx(),\n",
582 " 'Trendline' : MomentumFactors.make_trendline()\n",
583 "}"
584 ]
585 },
586 {
587 "cell_type": "code",
588 "execution_count": 15,
589 "metadata": {},
590 "outputs": [
591 {
592 "name": "stdout",
593 "output_type": "stream",
594 "text": [
595 "Pipeline run time 21.78 secs\n",
596 "<class 'pandas.core.frame.DataFrame'>\n",
597 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
598 "Data columns (total 6 columns):\n",
599 "Directional Movement Index 50362 non-null float64\n",
600 "Money Flow Index 50362 non-null float64\n",
601 "Percent Above Low 49536 non-null float64\n",
602 "Percent Below High 49536 non-null float64\n",
603 "Price Oscillator 50355 non-null float64\n",
604 "Trendline 49536 non-null float64\n",
605 "dtypes: float64(6)\n",
606 "memory usage: 2.7+ MB\n"
607 ]
608 }
609 ],
610 "source": [
611 "momentum_result, t = factor_pipeline(MOMENTUM_FACTORS)\n",
612 "print('Pipeline run time {:.2f} secs'.format(t))\n",
613 "momentum_result.info()"
614 ]
615 },
616 {
617 "cell_type": "markdown",
618 "metadata": {},
619 "source": [
620 "### Efficiency"
621 ]
622 },
623 {
624 "cell_type": "code",
625 "execution_count": 16,
626 "metadata": {},
627 "outputs": [],
628 "source": [
629 "class EfficiencyFactors:\n",
630 "\n",
631 " @staticmethod\n",
632 " def CapexToAssets(mask):\n",
633 " \"\"\"Capital Expenditure divided by Total Assets\"\"\"\n",
634 " capex = AnnualizedData(inputs = [Fundamentals.capital_expenditure_asof_date,\n",
635 " Fundamentals.capital_expenditure],\n",
636 " mask=mask) \n",
637 " assets = Fundamentals.total_assets.latest\n",
638 " return - capex / assets\n",
639 "\n",
640 " @staticmethod\n",
641 " def CapexToSales(mask):\n",
642 " \"\"\"Capital Expenditure divided by Total Revenue\"\"\"\n",
643 " capex = AnnualizedData(inputs = [Fundamentals.capital_expenditure_asof_date,\n",
644 " Fundamentals.capital_expenditure],\n",
645 " mask=mask) \n",
646 " revenue = AnnualizedData(inputs = [Fundamentals.total_revenue_asof_date,\n",
647 " Fundamentals.total_revenue],\n",
648 " mask=mask) \n",
649 " return - capex / revenue\n",
650 " \n",
651 " @staticmethod\n",
652 " def CapexToFCF(mask):\n",
653 " \"\"\"Capital Expenditure divided by Free Cash Flows\"\"\"\n",
654 " capex = AnnualizedData(inputs = [Fundamentals.capital_expenditure_asof_date,\n",
655 " Fundamentals.capital_expenditure],\n",
656 " mask=mask) \n",
657 " free_cash_flow = AnnualizedData(inputs = [Fundamentals.free_cash_flow_asof_date,\n",
658 " Fundamentals.free_cash_flow],\n",
659 " mask=mask) \n",
660 " return - capex / free_cash_flow\n",
661 "\n",
662 " @staticmethod\n",
663 " def EBITToAssets(mask):\n",
664 " \"\"\"Earnings Before Interest and Taxes (EBIT) divided by Total Assets\"\"\"\n",
665 " ebit = AnnualizedData(inputs = [Fundamentals.ebit_asof_date,\n",
666 " Fundamentals.ebit],\n",
667 " mask=mask) \n",
668 " assets = Fundamentals.total_assets.latest\n",
669 " return ebit / assets\n",
670 " \n",
671 " @staticmethod\n",
672 " def CFOToAssets(mask):\n",
673 " \"\"\"Operating Cash Flows divided by Total Assets\"\"\"\n",
674 " cfo = AnnualizedData(inputs = [Fundamentals.operating_cash_flow_asof_date,\n",
675 " Fundamentals.operating_cash_flow],\n",
676 " mask=mask) \n",
677 " assets = Fundamentals.total_assets.latest\n",
678 " return cfo / assets \n",
679 " \n",
680 " @staticmethod\n",
681 " def RetainedEarningsToAssets(mask):\n",
682 " \"\"\"Retained Earnings divided by Total Assets\"\"\"\n",
683 " retained_earnings = AnnualizedData(inputs = [Fundamentals.retained_earnings_asof_date,\n",
684 " Fundamentals.retained_earnings],\n",
685 " mask=mask) \n",
686 " assets = Fundamentals.total_assets.latest\n",
687 " return retained_earnings / assets"
688 ]
689 },
690 {
691 "cell_type": "code",
692 "execution_count": 17,
693 "metadata": {},
694 "outputs": [],
695 "source": [
696 "EFFICIENCY_FACTORS = {\n",
697 " 'CFO To Assets' :EfficiencyFactors.CFOToAssets,\n",
698 " 'Capex To Assets' :EfficiencyFactors.CapexToAssets,\n",
699 " 'Capex To FCF' :EfficiencyFactors.CapexToFCF,\n",
700 " 'Capex To Sales' :EfficiencyFactors.CapexToSales,\n",
701 " 'EBIT To Assets' :EfficiencyFactors.EBITToAssets,\n",
702 " 'Retained Earnings To Assets' :EfficiencyFactors.RetainedEarningsToAssets\n",
703 " }"
704 ]
705 },
706 {
707 "cell_type": "code",
708 "execution_count": 18,
709 "metadata": {},
710 "outputs": [
711 {
712 "name": "stdout",
713 "output_type": "stream",
714 "text": [
715 "Pipeline run time 37.93 secs\n",
716 "<class 'pandas.core.frame.DataFrame'>\n",
717 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
718 "Data columns (total 6 columns):\n",
719 "CFO To Assets 50351 non-null float64\n",
720 "Capex To Assets 46997 non-null float64\n",
721 "Capex To FCF 45799 non-null float64\n",
722 "Capex To Sales 46997 non-null float64\n",
723 "EBIT To Assets 46635 non-null float64\n",
724 "Retained Earnings To Assets 50349 non-null float64\n",
725 "dtypes: float64(6)\n",
726 "memory usage: 2.7+ MB\n"
727 ]
728 }
729 ],
730 "source": [
731 "efficiency_result, t = factor_pipeline(EFFICIENCY_FACTORS)\n",
732 "print('Pipeline run time {:.2f} secs'.format(t))\n",
733 "efficiency_result.info()"
734 ]
735 },
736 {
737 "cell_type": "markdown",
738 "metadata": {},
739 "source": [
740 "### Risk"
741 ]
742 },
743 {
744 "cell_type": "code",
745 "execution_count": 19,
746 "metadata": {},
747 "outputs": [],
748 "source": [
749 "class RiskFactors:\n",
750 "\n",
751 " @staticmethod\n",
752 " def LogMarketCap(mask):\n",
753 " \"\"\"Log of Market Capitalization log(Close Price * Shares Outstanding)\"\"\"\n",
754 " return np.log(MarketCap(mask=mask))\n",
755 " \n",
756 " class DownsideRisk(CustomFactor):\n",
757 " \"\"\"Mean returns divided by std of 1yr daily losses (Sortino Ratio)\"\"\"\n",
758 " inputs = [USEquityPricing.close]\n",
759 " window_length = 252\n",
760 "\n",
761 " def compute(self, today, assets, out, close):\n",
762 " ret = pd.DataFrame(close).pct_change()\n",
763 " out[:] = ret.mean().div(ret.where(ret<0).std())\n",
764 "\n",
765 " @staticmethod\n",
766 " def MarketBeta(**kwargs):\n",
767 " \"\"\"Slope of 1-yr regression of price returns against index returns\"\"\"\n",
768 " return SimpleBeta(target=symbols('SPY'), regression_length=252) \n",
769 "\n",
770 " class DownsideBeta(CustomFactor):\n",
771 " \"\"\"Slope of 1yr regression of returns on negative index returns\"\"\"\n",
772 " inputs = [USEquityPricing.close]\n",
773 " window_length = 252\n",
774 "\n",
775 " def compute(self, today, assets, out, close):\n",
776 " t = len(close)\n",
777 " assets = pd.DataFrame(close).pct_change()\n",
778 " \n",
779 " start_date = (today - pd.DateOffset(years=1)).strftime('%Y-%m-%d')\n",
780 " spy = get_pricing('SPY', \n",
781 " start_date=start_date, \n",
782 " end_date=today.strftime('%Y-%m-%d')).reset_index(drop=True)\n",
783 " spy_neg_ret = (spy\n",
784 " .close_price\n",
785 " .iloc[-t:]\n",
786 " .pct_change()\n",
787 " .pipe(lambda x: x.where(x<0)))\n",
788 " \n",
789 " out[:] = assets.apply(lambda x: x.cov(spy_neg_ret)).div(spy_neg_ret.var()) \n",
790 "\n",
791 " class Vol3M(CustomFactor):\n",
792 " \"\"\"3-month Volatility: Standard deviation of returns over 3 months\"\"\"\n",
793 "\n",
794 " inputs = [USEquityPricing.close]\n",
795 " window_length = 63\n",
796 "\n",
797 " def compute(self, today, assets, out, close):\n",
798 " out[:] = np.log1p(pd.DataFrame(close).pct_change()).std()"
799 ]
800 },
801 {
802 "cell_type": "code",
803 "execution_count": 20,
804 "metadata": {},
805 "outputs": [],
806 "source": [
807 "RISK_FACTORS = {\n",
808 " 'Log Market Cap' : RiskFactors.LogMarketCap,\n",
809 " 'Downside Risk' : RiskFactors.DownsideRisk,\n",
810 " 'Index Beta' : RiskFactors.MarketBeta,\n",
811 "# 'Downside Beta' : RiskFactors.DownsideBeta, \n",
812 " 'Volatility 3M' : RiskFactors.Vol3M, \n",
813 "}"
814 ]
815 },
816 {
817 "cell_type": "code",
818 "execution_count": 21,
819 "metadata": {},
820 "outputs": [
821 {
822 "name": "stdout",
823 "output_type": "stream",
824 "text": [
825 "Pipeline run time 48.59 secs\n",
826 "<class 'pandas.core.frame.DataFrame'>\n",
827 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
828 "Data columns (total 4 columns):\n",
829 "Downside Risk 50362 non-null float64\n",
830 "Index Beta 50079 non-null float64\n",
831 "Log Market Cap 50362 non-null float64\n",
832 "Volatility 3M 50362 non-null float64\n",
833 "dtypes: float64(4)\n",
834 "memory usage: 1.9+ MB\n"
835 ]
836 }
837 ],
838 "source": [
839 "risk_result, t = factor_pipeline(RISK_FACTORS)\n",
840 "print('Pipeline run time {:.2f} secs'.format(t))\n",
841 "risk_result.info()"
842 ]
843 },
844 {
845 "cell_type": "markdown",
846 "metadata": {},
847 "source": [
848 "### Growth"
849 ]
850 },
851 {
852 "cell_type": "code",
853 "execution_count": 22,
854 "metadata": {},
855 "outputs": [],
856 "source": [
857 "def growth_pipeline():\n",
858 " revenue = AnnualizedData(inputs = [Fundamentals.total_revenue_asof_date,\n",
859 " Fundamentals.total_revenue],\n",
860 " mask=UNIVERSE)\n",
861 " eps = AnnualizedData(inputs = [Fundamentals.diluted_eps_earnings_reports_asof_date,\n",
862 " Fundamentals.diluted_eps_earnings_reports],\n",
863 " mask=UNIVERSE) \n",
864 "\n",
865 " return Pipeline({'Sales': revenue,\n",
866 " 'EPS': eps,\n",
867 " 'Total Assets': Fundamentals.total_assets.latest,\n",
868 " 'Net Debt': Fundamentals.net_debt.latest},\n",
869 " screen=UNIVERSE)"
870 ]
871 },
872 {
873 "cell_type": "code",
874 "execution_count": 23,
875 "metadata": {},
876 "outputs": [
877 {
878 "name": "stdout",
879 "output_type": "stream",
880 "text": [
881 "Pipeline run time 23.48 secs\n",
882 "<class 'pandas.core.frame.DataFrame'>\n",
883 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
884 "Data columns (total 12 columns):\n",
885 "EPS 50215 non-null float64\n",
886 "Net Debt 47413 non-null float64\n",
887 "Sales 50351 non-null float64\n",
888 "Total Assets 50362 non-null float64\n",
889 "EPS Growth 3M 50152 non-null float64\n",
890 "EPS Growth 12M 49963 non-null float64\n",
891 "Net Debt Growth 3M 47350 non-null float64\n",
892 "Net Debt Growth 12M 47171 non-null float64\n",
893 "Sales Growth 3M 50288 non-null float64\n",
894 "Sales Growth 12M 50099 non-null float64\n",
895 "Total Assets Growth 3M 50299 non-null float64\n",
896 "Total Assets Growth 12M 50110 non-null float64\n",
897 "dtypes: float64(12)\n",
898 "memory usage: 5.0+ MB\n"
899 ]
900 }
901 ],
902 "source": [
903 "start_timer = time()\n",
904 "growth_result = run_pipeline(growth_pipeline(), start_date=START, end_date=END)\n",
905 "\n",
906 "for col in growth_result.columns:\n",
907 " for month in [3, 12]:\n",
908 " new_col = col + ' Growth {}M'.format(month)\n",
909 " kwargs = {new_col: growth_result[col].pct_change(month*MONTH).groupby(level=1).rank()} \n",
910 " growth_result = growth_result.assign(**kwargs)\n",
911 "print('Pipeline run time {:.2f} secs'.format(time() - start_timer))\n",
912 "growth_result.info()"
913 ]
914 },
915 {
916 "cell_type": "markdown",
917 "metadata": {},
918 "source": [
919 "### Quality"
920 ]
921 },
922 {
923 "cell_type": "code",
924 "execution_count": 24,
925 "metadata": {},
926 "outputs": [],
927 "source": [
928 "class QualityFactors:\n",
929 " \n",
930 " @staticmethod\n",
931 " def AssetTurnover(mask):\n",
932 " \"\"\"Sales divided by average of year beginning and year end assets\"\"\"\n",
933 "\n",
934 " assets = AnnualAvg(inputs=[Fundamentals.total_assets],\n",
935 " mask=mask)\n",
936 " sales = AnnualizedData([Fundamentals.total_revenue_asof_date,\n",
937 " Fundamentals.total_revenue], mask=mask)\n",
938 " return sales / assets\n",
939 " \n",
940 " @staticmethod\n",
941 " def CurrentRatio(mask):\n",
942 " \"\"\"Total current assets divided by total current liabilities\"\"\"\n",
943 "\n",
944 " assets = Fundamentals.current_assets.latest\n",
945 " liabilities = Fundamentals.current_liabilities.latest\n",
946 " return assets / liabilities\n",
947 " \n",
948 " @staticmethod\n",
949 " def AssetToEquityRatio(mask):\n",
950 " \"\"\"Total current assets divided by common equity\"\"\"\n",
951 "\n",
952 " assets = Fundamentals.current_assets.latest\n",
953 " equity = Fundamentals.common_stock.latest\n",
954 " return assets / equity \n",
955 "\n",
956 " \n",
957 " @staticmethod\n",
958 " def InterestCoverage(mask):\n",
959 " \"\"\"EBIT divided by interest expense\"\"\"\n",
960 "\n",
961 " ebit = AnnualizedData(inputs = [Fundamentals.ebit_asof_date,\n",
962 " Fundamentals.ebit], mask=mask) \n",
963 " \n",
964 " interest_expense = AnnualizedData(inputs = [Fundamentals.interest_expense_asof_date,\n",
965 " Fundamentals.interest_expense], mask=mask)\n",
966 " return ebit / interest_expense\n",
967 "\n",
968 " @staticmethod\n",
969 " def DebtToAssetRatio(mask):\n",
970 " \"\"\"Total Debts divided by Total Assets\"\"\"\n",
971 "\n",
972 " debt = Fundamentals.total_debt.latest\n",
973 " assets = Fundamentals.total_assets.latest\n",
974 " return debt / assets\n",
975 " \n",
976 " @staticmethod\n",
977 " def DebtToEquityRatio(mask):\n",
978 " \"\"\"Total Debts divided by Common Stock Equity\"\"\"\n",
979 "\n",
980 " debt = Fundamentals.total_debt.latest\n",
981 " equity = Fundamentals.common_stock.latest\n",
982 " return debt / equity \n",
983 "\n",
984 " @staticmethod\n",
985 " def WorkingCapitalToAssets(mask):\n",
986 " \"\"\"Current Assets less Current liabilities (Working Capital) divided by Assets\"\"\"\n",
987 "\n",
988 " working_capital = Fundamentals.working_capital.latest\n",
989 " assets = Fundamentals.total_assets.latest\n",
990 " return working_capital / assets\n",
991 " \n",
992 " @staticmethod\n",
993 " def WorkingCapitalToSales(mask):\n",
994 " \"\"\"Current Assets less Current liabilities (Working Capital), divided by Sales\"\"\"\n",
995 "\n",
996 " working_capital = Fundamentals.working_capital.latest\n",
997 " sales = AnnualizedData([Fundamentals.total_revenue_asof_date,\n",
998 " Fundamentals.total_revenue], mask=mask) \n",
999 " return working_capital / sales \n",
1000 " \n",
1001 " \n",
1002 " class MertonsDD(CustomFactor):\n",
1003 " \"\"\"Merton's Distance to Default \"\"\"\n",
1004 " \n",
1005 " inputs = [Fundamentals.total_assets,\n",
1006 " Fundamentals.total_liabilities, \n",
1007 " libor.value, \n",
1008 " USEquityPricing.close]\n",
1009 " window_length = 252\n",
1010 "\n",
1011 " def compute(self, today, assets, out, tot_assets, tot_liabilities, r, close):\n",
1012 " mertons = []\n",
1013 "\n",
1014 " for col_assets, col_liabilities, col_r, col_close in zip(tot_assets.T, tot_liabilities.T,\n",
1015 " r.T, close.T):\n",
1016 " vol_1y = np.nanstd(col_close)\n",
1017 " numerator = np.log(\n",
1018 " col_assets[-1] / col_liabilities[-1]) + ((252 * col_r[-1]) - ((vol_1y ** 2) / 2))\n",
1019 " mertons.append(numerator / vol_1y)\n",
1020 "\n",
1021 " out[:] = mertons "
1022 ]
1023 },
1024 {
1025 "cell_type": "code",
1026 "execution_count": 25,
1027 "metadata": {
1028 "scrolled": true
1029 },
1030 "outputs": [],
1031 "source": [
1032 "QUALITY_FACTORS = {\n",
1033 " 'AssetToEquityRatio' : QualityFactors.AssetToEquityRatio,\n",
1034 " 'AssetTurnover' : QualityFactors.AssetTurnover,\n",
1035 " 'CurrentRatio' : QualityFactors.CurrentRatio,\n",
1036 " 'DebtToAssetRatio' : QualityFactors.DebtToAssetRatio,\n",
1037 " 'DebtToEquityRatio' : QualityFactors.DebtToEquityRatio,\n",
1038 " 'InterestCoverage' : QualityFactors.InterestCoverage,\n",
1039 " 'MertonsDD' : QualityFactors.MertonsDD,\n",
1040 " 'WorkingCapitalToAssets': QualityFactors.WorkingCapitalToAssets,\n",
1041 " 'WorkingCapitalToSales' : QualityFactors.WorkingCapitalToSales,\n",
1042 "}\n",
1043 " "
1044 ]
1045 },
1046 {
1047 "cell_type": "code",
1048 "execution_count": 26,
1049 "metadata": {},
1050 "outputs": [
1051 {
1052 "name": "stdout",
1053 "output_type": "stream",
1054 "text": [
1055 "Pipeline run time 36.81 secs\n",
1056 "<class 'pandas.core.frame.DataFrame'>\n",
1057 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1058 "Data columns (total 9 columns):\n",
1059 "AssetToEquityRatio 45176 non-null float64\n",
1060 "AssetTurnover 50314 non-null float64\n",
1061 "CurrentRatio 45680 non-null float64\n",
1062 "DebtToAssetRatio 50080 non-null float64\n",
1063 "DebtToEquityRatio 48492 non-null float64\n",
1064 "InterestCoverage 35250 non-null float64\n",
1065 "MertonsDD 50362 non-null float64\n",
1066 "WorkingCapitalToAssets 45680 non-null float64\n",
1067 "WorkingCapitalToSales 45669 non-null float64\n",
1068 "dtypes: float64(9)\n",
1069 "memory usage: 3.8+ MB\n"
1070 ]
1071 }
1072 ],
1073 "source": [
1074 "quality_result, t = factor_pipeline(QUALITY_FACTORS)\n",
1075 "print('Pipeline run time {:.2f} secs'.format(t))\n",
1076 "quality_result.info()"
1077 ]
1078 },
1079 {
1080 "cell_type": "markdown",
1081 "metadata": {},
1082 "source": [
1083 "### Payout"
1084 ]
1085 },
1086 {
1087 "cell_type": "code",
1088 "execution_count": 27,
1089 "metadata": {},
1090 "outputs": [],
1091 "source": [
1092 "class PayoutFactors:\n",
1093 "\n",
1094 " @staticmethod\n",
1095 " def DividendPayoutRatio(mask):\n",
1096 " \"\"\"Dividends Per Share divided by Earnings Per Share\"\"\"\n",
1097 "\n",
1098 " dps = AnnualizedData(inputs = [Fundamentals.dividend_per_share_earnings_reports_asof_date,\n",
1099 " Fundamentals.dividend_per_share_earnings_reports], mask=mask) \n",
1100 " \n",
1101 " eps = AnnualizedData(inputs = [Fundamentals.basic_eps_earnings_reports_asof_date,\n",
1102 " Fundamentals.basic_eps_earnings_reports], mask=mask)\n",
1103 " return dps / eps\n",
1104 " \n",
1105 " @staticmethod\n",
1106 " def DividendGrowth(**kwargs):\n",
1107 " \"\"\"Annualized percentage DPS change\"\"\" \n",
1108 " return Fundamentals.dps_growth.latest "
1109 ]
1110 },
1111 {
1112 "cell_type": "code",
1113 "execution_count": 28,
1114 "metadata": {},
1115 "outputs": [],
1116 "source": [
1117 "PAYOUT_FACTORS = {\n",
1118 " 'Dividend Payout Ratio': PayoutFactors.DividendPayoutRatio,\n",
1119 " 'Dividend Growth': PayoutFactors.DividendGrowth\n",
1120 "}"
1121 ]
1122 },
1123 {
1124 "cell_type": "code",
1125 "execution_count": 29,
1126 "metadata": {},
1127 "outputs": [
1128 {
1129 "name": "stdout",
1130 "output_type": "stream",
1131 "text": [
1132 "Pipeline run time 23.15 secs\n",
1133 "<class 'pandas.core.frame.DataFrame'>\n",
1134 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1135 "Data columns (total 2 columns):\n",
1136 "Dividend Growth 40517 non-null float64\n",
1137 "Dividend Payout Ratio 39947 non-null float64\n",
1138 "dtypes: float64(2)\n",
1139 "memory usage: 1.2+ MB\n"
1140 ]
1141 }
1142 ],
1143 "source": [
1144 "payout_result, t = factor_pipeline(PAYOUT_FACTORS)\n",
1145 "print('Pipeline run time {:.2f} secs'.format(t))\n",
1146 "payout_result.info()"
1147 ]
1148 },
1149 {
1150 "cell_type": "markdown",
1151 "metadata": {},
1152 "source": [
1153 "### Profitability"
1154 ]
1155 },
1156 {
1157 "cell_type": "code",
1158 "execution_count": 30,
1159 "metadata": {},
1160 "outputs": [],
1161 "source": [
1162 "class ProfitabilityFactors:\n",
1163 " \n",
1164 " @staticmethod\n",
1165 " def GrossProfitMargin(mask):\n",
1166 " \"\"\"Gross Profit divided by Net Sales\"\"\"\n",
1167 "\n",
1168 " gross_profit = AnnualizedData([Fundamentals.gross_profit_asof_date,\n",
1169 " Fundamentals.gross_profit], mask=mask) \n",
1170 " sales = AnnualizedData([Fundamentals.total_revenue_asof_date,\n",
1171 " Fundamentals.total_revenue], mask=mask)\n",
1172 " return gross_profit / sales \n",
1173 " \n",
1174 " @staticmethod\n",
1175 " def NetIncomeMargin(mask):\n",
1176 " \"\"\"Net income divided by Net Sales\"\"\"\n",
1177 "\n",
1178 " net_income = AnnualizedData([Fundamentals.net_income_income_statement_asof_date,\n",
1179 " Fundamentals.net_income_income_statement], mask=mask) \n",
1180 " sales = AnnualizedData([Fundamentals.total_revenue_asof_date,\n",
1181 " Fundamentals.total_revenue], mask=mask)\n",
1182 " return net_income / sales "
1183 ]
1184 },
1185 {
1186 "cell_type": "code",
1187 "execution_count": 31,
1188 "metadata": {},
1189 "outputs": [],
1190 "source": [
1191 "PROFITABIILTY_FACTORS = {\n",
1192 " 'Gross Profit Margin': ProfitabilityFactors.GrossProfitMargin,\n",
1193 " 'Net Income Margin': ProfitabilityFactors.NetIncomeMargin,\n",
1194 " 'Return on Equity': Fundamentals.roe.latest,\n",
1195 " 'Return on Assets': Fundamentals.roa.latest,\n",
1196 " 'Return on Invested Capital': Fundamentals.roic.latest\n",
1197 "}"
1198 ]
1199 },
1200 {
1201 "cell_type": "code",
1202 "execution_count": 32,
1203 "metadata": {},
1204 "outputs": [
1205 {
1206 "name": "stdout",
1207 "output_type": "stream",
1208 "text": [
1209 "Pipeline run time 23.27 secs\n",
1210 "<class 'pandas.core.frame.DataFrame'>\n",
1211 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1212 "Data columns (total 2 columns):\n",
1213 "Dividend Growth 40517 non-null float64\n",
1214 "Dividend Payout Ratio 39947 non-null float64\n",
1215 "dtypes: float64(2)\n",
1216 "memory usage: 1.2+ MB\n"
1217 ]
1218 }
1219 ],
1220 "source": [
1221 "profitability_result, t = factor_pipeline(PAYOUT_FACTORS)\n",
1222 "print('Pipeline run time {:.2f} secs'.format(t))\n",
1223 "payout_result.info()"
1224 ]
1225 },
1226 {
1227 "cell_type": "code",
1228 "execution_count": 33,
1229 "metadata": {},
1230 "outputs": [],
1231 "source": [
1232 "# profitability_pipeline().show_graph(format='png')"
1233 ]
1234 },
1235 {
1236 "cell_type": "markdown",
1237 "metadata": {},
1238 "source": [
1239 "### Build Dataset"
1240 ]
1241 },
1242 {
1243 "cell_type": "markdown",
1244 "metadata": {},
1245 "source": [
1246 "#### Get Returns"
1247 ]
1248 },
1249 {
1250 "cell_type": "code",
1251 "execution_count": 34,
1252 "metadata": {},
1253 "outputs": [
1254 {
1255 "name": "stdout",
1256 "output_type": "stream",
1257 "text": [
1258 "<class 'pandas.core.frame.DataFrame'>\n",
1259 "MultiIndex: 50362 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1260 "Data columns (total 4 columns):\n",
1261 "Returns10D 50362 non-null float64\n",
1262 "Returns1D 50362 non-null float64\n",
1263 "Returns20D 50360 non-null float64\n",
1264 "Returns5D 50362 non-null float64\n",
1265 "dtypes: float64(4)\n",
1266 "memory usage: 1.9+ MB\n"
1267 ]
1268 }
1269 ],
1270 "source": [
1271 "lookahead = [1, 5, 10, 20]\n",
1272 "returns = run_pipeline(Pipeline({'Returns{}D'.format(i): Returns(inputs=[USEquityPricing.close], \n",
1273 " window_length=i+1, mask=UNIVERSE) for i in lookahead},\n",
1274 " screen=UNIVERSE),\n",
1275 " start_date=START, \n",
1276 " end_date=END)\n",
1277 "return_cols = ['Returns{}D'.format(i) for i in lookahead]\n",
1278 "returns.info()"
1279 ]
1280 },
1281 {
1282 "cell_type": "code",
1283 "execution_count": 35,
1284 "metadata": {},
1285 "outputs": [],
1286 "source": [
1287 "data = pd.concat([returns,\n",
1288 " value_result,\n",
1289 " momentum_result,\n",
1290 " quality_result,\n",
1291 " payout_result,\n",
1292 " growth_result,\n",
1293 " efficiency_result,\n",
1294 " risk_result], axis=1).sortlevel()\n",
1295 "data.index.names = ['date', 'asset']"
1296 ]
1297 },
1298 {
1299 "cell_type": "code",
1300 "execution_count": 36,
1301 "metadata": {},
1302 "outputs": [],
1303 "source": [
1304 "data['stock'] = data.index.get_level_values('asset').map(lambda x: x.asset_name)"
1305 ]
1306 },
1307 {
1308 "cell_type": "markdown",
1309 "metadata": {},
1310 "source": [
1311 "#### Remove columns and rows with less than 80% of data availability"
1312 ]
1313 },
1314 {
1315 "cell_type": "code",
1316 "execution_count": 37,
1317 "metadata": {},
1318 "outputs": [
1319 {
1320 "name": "stdout",
1321 "output_type": "stream",
1322 "text": [
1323 "2,985 rows and 3 columns dropped\n"
1324 ]
1325 }
1326 ],
1327 "source": [
1328 "rows_before, cols_before = data.shape\n",
1329 "data = (data\n",
1330 " .dropna(axis=1, thresh=int(len(data)*.8))\n",
1331 " .dropna(thresh=int(len(data.columns) * .8)))\n",
1332 "data = data.fillna(data.median())\n",
1333 "rows_after, cols_after = data.shape\n",
1334 "print('{:,d} rows and {:,d} columns dropped'.format(rows_before-rows_after, cols_before-cols_after))"
1335 ]
1336 },
1337 {
1338 "cell_type": "code",
1339 "execution_count": 38,
1340 "metadata": {},
1341 "outputs": [
1342 {
1343 "name": "stdout",
1344 "output_type": "stream",
1345 "text": [
1346 "<class 'pandas.core.frame.DataFrame'>\n",
1347 "MultiIndex: 47377 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1348 "Data columns (total 52 columns):\n",
1349 "AssetToEquityRatio 47377 non-null float64\n",
1350 "AssetTurnover 47377 non-null float64\n",
1351 "CFO To Assets 47377 non-null float64\n",
1352 "Capex To Assets 47377 non-null float64\n",
1353 "Capex To FCF 47377 non-null float64\n",
1354 "Capex To Sales 47377 non-null float64\n",
1355 "CurrentRatio 47377 non-null float64\n",
1356 "DebtToAssetRatio 47377 non-null float64\n",
1357 "DebtToEquityRatio 47377 non-null float64\n",
1358 "Directional Movement Index 47377 non-null float64\n",
1359 "Dividend Growth 47377 non-null float64\n",
1360 "DividendYield 47377 non-null float64\n",
1361 "Downside Risk 47377 non-null float64\n",
1362 "EBIT To Assets 47377 non-null float64\n",
1363 "EBITDAYield 47377 non-null float64\n",
1364 "EPS 47377 non-null float64\n",
1365 "EPS Growth 12M 47377 non-null float64\n",
1366 "EPS Growth 3M 47377 non-null float64\n",
1367 "EVToEBITDA 47377 non-null float64\n",
1368 "EVToFCF 47377 non-null float64\n",
1369 "Index Beta 47377 non-null float64\n",
1370 "Log Market Cap 47377 non-null float64\n",
1371 "MertonsDD 47377 non-null float64\n",
1372 "Money Flow Index 47377 non-null float64\n",
1373 "Net Debt 47377 non-null float64\n",
1374 "Net Debt Growth 12M 47377 non-null float64\n",
1375 "Net Debt Growth 3M 47377 non-null float64\n",
1376 "Percent Above Low 47377 non-null float64\n",
1377 "Percent Below High 47377 non-null float64\n",
1378 "Price Oscillator 47377 non-null float64\n",
1379 "PriceToBook 47377 non-null float64\n",
1380 "PriceToDilutedEarningsTTM 47377 non-null float64\n",
1381 "PriceToEarningsTTM 47377 non-null float64\n",
1382 "PriceToFCF 47377 non-null float64\n",
1383 "PriceToOperatingCashflow 47377 non-null float64\n",
1384 "PriceToSalesTTM 47377 non-null float64\n",
1385 "Retained Earnings To Assets 47377 non-null float64\n",
1386 "Returns10D 47377 non-null float64\n",
1387 "Returns1D 47377 non-null float64\n",
1388 "Returns20D 47377 non-null float64\n",
1389 "Returns5D 47377 non-null float64\n",
1390 "Sales 47377 non-null float64\n",
1391 "Sales Growth 12M 47377 non-null float64\n",
1392 "Sales Growth 3M 47377 non-null float64\n",
1393 "Total Assets 47377 non-null float64\n",
1394 "Total Assets Growth 12M 47377 non-null float64\n",
1395 "Total Assets Growth 3M 47377 non-null float64\n",
1396 "Trendline 47377 non-null float64\n",
1397 "Volatility 3M 47377 non-null float64\n",
1398 "WorkingCapitalToAssets 47377 non-null float64\n",
1399 "WorkingCapitalToSales 47377 non-null float64\n",
1400 "stock 47377 non-null object\n",
1401 "dtypes: float64(51), object(1)\n",
1402 "memory usage: 19.2+ MB\n"
1403 ]
1404 }
1405 ],
1406 "source": [
1407 "data.sort_index(1).info()"
1408 ]
1409 },
1410 {
1411 "cell_type": "code",
1412 "execution_count": 39,
1413 "metadata": {},
1414 "outputs": [
1415 {
1416 "data": {
1417 "image/png": "iVBORw0KGgoAAAANSUhEUgAAAywAAAOjCAYAAABZaICQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
1418 "text/plain": [
1419 "<matplotlib.figure.Figure at 0x7feb6b7d1e50>"
1420 ]
1421 },
1422 "metadata": {},
1423 "output_type": "display_data"
1424 }
1425 ],
1426 "source": [
1427 "g = sns.clustermap(data.drop(['stock'] + return_cols, axis=1).corr())\n",
1428 "plt.gcf().set_size_inches((14,14));"
1429 ]
1430 },
1431 {
1432 "cell_type": "markdown",
1433 "metadata": {},
1434 "source": [
1435 "#### Prepare Features"
1436 ]
1437 },
1438 {
1439 "cell_type": "code",
1440 "execution_count": 40,
1441 "metadata": {},
1442 "outputs": [
1443 {
1444 "name": "stdout",
1445 "output_type": "stream",
1446 "text": [
1447 "<class 'pandas.core.frame.DataFrame'>\n",
1448 "MultiIndex: 47377 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1449 "Columns: 182 entries, DividendYield to stock_YELP INC\n",
1450 "dtypes: float64(182)\n",
1451 "memory usage: 66.1+ MB\n"
1452 ]
1453 }
1454 ],
1455 "source": [
1456 "X = pd.get_dummies(data.drop(return_cols, axis=1))\n",
1457 "X.info()"
1458 ]
1459 },
1460 {
1461 "cell_type": "markdown",
1462 "metadata": {},
1463 "source": [
1464 "#### Shifted Returns"
1465 ]
1466 },
1467 {
1468 "cell_type": "code",
1469 "execution_count": 41,
1470 "metadata": {
1471 "scrolled": true
1472 },
1473 "outputs": [
1474 {
1475 "name": "stdout",
1476 "output_type": "stream",
1477 "text": [
1478 "<class 'pandas.core.frame.DataFrame'>\n",
1479 "MultiIndex: 47377 entries, (2014-01-02 00:00:00+00:00, Equity(24 [AAPL])) to (2015-12-31 00:00:00+00:00, Equity(47208 [GPRO]))\n",
1480 "Data columns (total 4 columns):\n",
1481 "Returns1D 47242 non-null float64\n",
1482 "Returns5D 46706 non-null float64\n",
1483 "Returns10D 46036 non-null float64\n",
1484 "Returns20D 44696 non-null float64\n",
1485 "dtypes: float64(4)\n",
1486 "memory usage: 1.8+ MB\n"
1487 ]
1488 }
1489 ],
1490 "source": [
1491 "y = data.loc[:, return_cols]\n",
1492 "shifted_y = []\n",
1493 "for col in y.columns:\n",
1494 " t = int(re.search(r'\\d+', col).group(0))\n",
1495 " shifted_y.append(y.groupby(level='asset')['Returns{}D'.format(t)].shift(-t).to_frame(col))\n",
1496 "y = pd.concat(shifted_y, axis=1)\n",
1497 "y.info()"
1498 ]
1499 },
1500 {
1501 "cell_type": "code",
1502 "execution_count": 42,
1503 "metadata": {},
1504 "outputs": [
1505 {
1506 "name": "stderr",
1507 "output_type": "stream",
1508 "text": [
1509 "/usr/local/lib/python2.7/dist-packages/seaborn/categorical.py:2171: UserWarning: The boxplot API has been changed. Attempting to adjust your arguments for the new API (which might not work). Please update your code. See the version 0.6 release notes for more info.\n",
1510 " warnings.warn(msg, UserWarning)\n"
1511 ]
1512 },
1513 {
1514 "data": {
1515 "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzYAAAHrCAYAAAAdcAiZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwQYAAACA7f0/9mHOqH2g4isAAAAASUVORK5CYII=\n",
1516 "text/plain": [
1517 "<matplotlib.figure.Figure at 0x7feb6c0bfed0>"
1518 ]
1519 },
1520 "metadata": {},
1521 "output_type": "display_data"
1522 }
1523 ],
1524 "source": [
1525 "ax = sns.boxplot(y[return_cols])\n",
1526 "ax.set_title('Return Distriubtions');"
1527 ]
1528 },
1529 {
1530 "cell_type": "code",
1531 "execution_count": 44,
1532 "metadata": {},
1533 "outputs": [
1534 {
1535 "name": "stdout",
1536 "output_type": "stream",
1537 "text": [
1538 "<class 'pandas.core.frame.DataFrame'>\n",
1539 "DatetimeIndex: 45114 entries, 2014-01-02 to 2015-12-16\n",
1540 "Columns: 183 entries, Returns10D to stock_YELP INC\n",
1541 "dtypes: float64(183)\n",
1542 "memory usage: 63.3 MB\n",
1543 "None\n"
1544 ]
1545 }
1546 ],
1547 "source": [
1548 "target = 'Returns10D'\n",
1549 "outliers = .01\n",
1550 "model_data = pd.concat([y[[target]], X], axis=1).dropna().reset_index('asset', drop=True)\n",
1551 "model_data = model_data[model_data[target].between(*model_data[target].quantile([outliers, 1-outliers]).values)] \n",
1552 "\n",
1553 "model_data[target] = np.log1p(model_data[target])\n",
1554 "features = model_data.drop(target, axis=1).columns\n",
1555 "dates = model_data.index.unique()\n",
1556 "\n",
1557 "print(model_data.info())"
1558 ]
1559 },
1560 {
1561 "cell_type": "code",
1562 "execution_count": 45,
1563 "metadata": {},
1564 "outputs": [
1565 {
1566 "data": {
1567 "text/plain": [
1568 "count 45114.000000\n",
1569 "mean 0.001159\n",
1570 "std 0.045740\n",
1571 "min -0.157448\n",
1572 "25% -0.025013\n",
1573 "50% 0.002817\n",
1574 "75% 0.028880\n",
1575 "max 0.146139\n",
1576 "Name: Returns10D, dtype: float64"
1577 ]
1578 },
1579 "execution_count": 45,
1580 "metadata": {},
1581 "output_type": "execute_result"
1582 }
1583 ],
1584 "source": [
1585 "model_data[target].describe()"
1586 ]
1587 },
1588 {
1589 "cell_type": "code",
1590 "execution_count": 46,
1591 "metadata": {},
1592 "outputs": [],
1593 "source": [
1594 "idx = pd.IndexSlice"
1595 ]
1596 },
1597 {
1598 "cell_type": "markdown",
1599 "metadata": {},
1600 "source": [
1601 "### Logistic Regression: Classification"
1602 ]
1603 },
1604 {
1605 "cell_type": "code",
1606 "execution_count": 49,
1607 "metadata": {},
1608 "outputs": [],
1609 "source": [
1610 "def time_series_split(d, nfolds=5, min_train=21):\n",
1611 " \"\"\"Generate train/test dates for nfolds \n",
1612 " with at least min_train train obs\n",
1613 " \"\"\"\n",
1614 " train_dates = d[:min_train].tolist()\n",
1615 " n = int(len(dates)/(nfolds + 1)) + 1\n",
1616 " test_folds = [d[i:i + n] for i in range(min_train, len(d), n)]\n",
1617 " for test_dates in test_folds:\n",
1618 " if len(train_dates) > min_train:\n",
1619 " yield train_dates, test_dates\n",
1620 " train_dates.extend(test_dates)"
1621 ]
1622 },
1623 {
1624 "cell_type": "code",
1625 "execution_count": 47,
1626 "metadata": {},
1627 "outputs": [
1628 {
1629 "name": "stdout",
1630 "output_type": "stream",
1631 "text": [
1632 "<class 'pandas.core.frame.DataFrame'>\n",
1633 "DatetimeIndex: 47377 entries, 2014-01-02 to 2015-12-31\n",
1634 "Columns: 183 entries, Returns10D to stock_YELP INC\n",
1635 "dtypes: float64(182), int64(1)\n",
1636 "memory usage: 66.5 MB\n",
1637 "None\n"
1638 ]
1639 }
1640 ],
1641 "source": [
1642 "target = 'Returns10D'\n",
1643 "label = (y[target] > 0).astype(int).to_frame(target)\n",
1644 "model_data = pd.concat([label, X], axis=1).dropna().reset_index('asset', drop=True)\n",
1645 "\n",
1646 "features = model_data.drop(target, axis=1).columns\n",
1647 "dates = model_data.index.unique()\n",
1648 "\n",
1649 "print(model_data.info())"
1650 ]
1651 },
1652 {
1653 "cell_type": "markdown",
1654 "metadata": {},
1655 "source": [
1656 "With this new categorical outcome variable, we can now train a logistic regression using the default L2 regularization. For logistic regression, the regularization is formulated inversely to linear regression: higher values for λ imply less regularization and vice versa. We evaluate 11 parameter values using cross validation as follows:"
1657 ]
1658 },
1659 {
1660 "cell_type": "code",
1661 "execution_count": 62,
1662 "metadata": {},
1663 "outputs": [
1664 {
1665 "name": "stdout",
1666 "output_type": "stream",
1667 "text": [
1668 "1e-05\n",
1669 "0.0001\n",
1670 "0.001\n",
1671 "0.01\n",
1672 "0.1\n",
1673 "1.0\n",
1674 "10.0\n",
1675 "100.0\n",
1676 "1000.0\n",
1677 "10000.0\n",
1678 "100000.0\n"
1679 ]
1680 }
1681 ],
1682 "source": [
1683 "nfolds = 250\n",
1684 "Cs = np.logspace(-5, 5, 11)\n",
1685 "scaler = StandardScaler()\n",
1686 "\n",
1687 "logistic_results, logistic_coeffs = pd.DataFrame(), pd.DataFrame()\n",
1688 "for C in Cs:\n",
1689 " print(C)\n",
1690 " coeffs = []\n",
1691 " log_reg = LogisticRegression(C=C)\n",
1692 " for i, (train_dates, test_dates) in enumerate(time_series_split(dates, nfolds=nfolds)):\n",
1693 "\n",
1694 " X_train = model_data.loc[idx[train_dates], features]\n",
1695 " y_train = model_data.loc[idx[train_dates], target]\n",
1696 " log_reg.fit(X=scaler.fit_transform(X_train), y=y_train)\n",
1697 "\n",
1698 " X_test = model_data.loc[idx[test_dates], features]\n",
1699 " y_test = model_data.loc[idx[test_dates], target]\n",
1700 " y_pred = log_reg.predict_proba(scaler.transform(X_test))[:, 1]\n",
1701 " \n",
1702 " coeffs.append(log_reg.coef_.squeeze())\n",
1703 " logistic_results = (logistic_results\n",
1704 " .append(y_test\n",
1705 " .to_frame('actuals')\n",
1706 " .assign(predicted=y_pred, C=C)))\n",
1707 " logistic_coeffs[C] = np.mean(coeffs, axis=0)"
1708 ]
1709 },
1710 {
1711 "cell_type": "markdown",
1712 "metadata": {},
1713 "source": [
1714 "### Evaluate Results using AUC Score"
1715 ]
1716 },
1717 {
1718 "cell_type": "markdown",
1719 "metadata": {},
1720 "source": [
1721 "We then use the roc_auc_score discussed in the previous chapter to compare the predictive accuracy across the various regularization parameters:"
1722 ]
1723 },
1724 {
1725 "cell_type": "code",
1726 "execution_count": 84,
1727 "metadata": {},
1728 "outputs": [],
1729 "source": [
1730 "auc_by_C = logistic_results.groupby('C').apply(lambda x: roc_auc_score(y_true=x.actuals.astype(int), \n",
1731 " y_score=x.predicted))"
1732 ]
1733 },
1734 {
1735 "cell_type": "code",
1736 "execution_count": 85,
1737 "metadata": {
1738 "scrolled": true
1739 },
1740 "outputs": [
1741 {
1742 "data": {
1743 "text/plain": [
1744 "C\n",
1745 "0.00001 0.532385\n",
1746 "0.00010 0.539814\n",
1747 "0.00100 0.549100\n",
1748 "0.01000 0.554121\n",
1749 "0.10000 0.561894\n",
1750 "1.00000 0.569875\n",
1751 "10.00000 0.574998\n",
1752 "100.00000 0.575875\n",
1753 "1000.00000 0.575214\n",
1754 "10000.00000 0.574488\n",
1755 "100000.00000 0.574317\n",
1756 "dtype: float64"
1757 ]
1758 },
1759 "execution_count": 85,
1760 "metadata": {},
1761 "output_type": "execute_result"
1762 }
1763 ],
1764 "source": [
1765 "auc_by_C"
1766 ]
1767 },
1768 {
1769 "cell_type": "markdown",
1770 "metadata": {},
1771 "source": [
1772 "We can again plot the AUC result for the range of hyperparameter values alongside the coefficient path that shows the improvements in predictive accuracy as the coefficients are a bit shrunk at the optimal regularization value 102:"
1773 ]
1774 },
1775 {
1776 "cell_type": "code",
1777 "execution_count": 86,
1778 "metadata": {},
1779 "outputs": [],
1780 "source": [
1781 "base_auc = auc.iloc[-1]\n",
1782 "best_auc = auc.max()\n",
1783 "best_C = auc.idxmax()"
1784 ]
1785 },
1786 {
1787 "cell_type": "code",
1788 "execution_count": 88,
1789 "metadata": {},
1790 "outputs": [
1791 {
1792 "data": {
1793 "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+QAAAI0CAYAAACUFOOKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nADCAgRwAAAAAAAP+f35RaXece/lXAAAAAElFTkSuQmCC\n",
1794 "text/plain": [
1795 "<matplotlib.figure.Figure at 0x7feb67a2b610>"
1796 ]
1797 },
1798 "metadata": {},
1799 "output_type": "display_data"
1800 }
1801 ],
1802 "source": [
1803 "fig, axes = plt.subplots(ncols=2, sharex=True)\n",
1804 "\n",
1805 "auc_by_C.sort_index(ascending=False).plot(logx=True, title='Area under the Curve (AUC)', ax=axes[0])\n",
1806 "axes[0].axhline(y=base_auc, c='black')\n",
1807 "axes[0].axvline(x=best_C, c='darkgrey', ls='--')\n",
1808 "axes[0].set_xlabel('Regularization')\n",
1809 "axes[0].set_ylabel('Information Coefficient')\n",
1810 "\n",
1811 "logistic_coeffs.T.sort_index(ascending=False).plot(legend=False, logx=True, title='Logistic Ridge Path', ax=axes[1])\n",
1812 "axes[1].set_xlabel('Regularization')\n",
1813 "axes[1].set_ylabel('Coefficients')\n",
1814 "axes[1].axvline(x=best_C, c='darkgrey', ls='--')\n",
1815 "fig.tight_layout();"
1816 ]
1817 },
1818 {
1819 "cell_type": "markdown",
1820 "metadata": {},
1821 "source": [
1822 "### Ordinal Logit"
1823 ]
1824 },
1825 {
1826 "cell_type": "code",
1827 "execution_count": null,
1828 "metadata": {},
1829 "outputs": [],
1830 "source": [
1831 "target = 'Returns10D'\n",
1832 "label = (y[target] > 0).astype(int).to_frame(target)\n",
1833 "model_data = pd.concat([label, X], axis=1).dropna().reset_index('asset', drop=True)\n",
1834 "\n",
1835 "features = model_data.drop(target, axis=1).columns\n",
1836 "dates = model_data.index.unique()\n",
1837 "\n",
1838 "print(model_data.info())"
1839 ]
1840 },
1841 {
1842 "cell_type": "code",
1843 "execution_count": null,
1844 "metadata": {},
1845 "outputs": [],
1846 "source": []
1847 }
1848 ],
1849 "metadata": {
1850 "kernelspec": {
1851 "display_name": "Python 2",
1852 "language": "python",
1853 "name": "python2"
1854 },
1855 "language_info": {
1856 "codemirror_mode": {
1857 "name": "ipython",
1858 "version": 3
1859 },
1860 "file_extension": ".py",
1861 "mimetype": "text/x-python",
1862 "name": "python",
1863 "nbconvert_exporter": "python",
1864 "pygments_lexer": "ipython3",
1865 "version": "3.7.0"
1866 },
1867 "toc": {
1868 "base_numbering": 1,
1869 "nav_menu": {},
1870 "number_sections": true,
1871 "sideBar": true,
1872 "skip_h1_title": true,
1873 "title_cell": "Table of Contents",
1874 "title_sidebar": "Contents",
1875 "toc_cell": false,
1876 "toc_position": {},
1877 "toc_section_display": true,
1878 "toc_window_display": true
1879 }
1880 },
1881 "nbformat": 4,
1882 "nbformat_minor": 2
1883 }