ml-finance-python

notebook.ipynb

(185163B)

 {
  "cells": [
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "# Autoregressive (AR) Models\n",
     "\n",
     "by Maxwell Margenot, Delaney Mackenzie, and Lee Tobey\n",
     "\n",
     "Lee Tobey is the founder of [Hedgewise](https://www.hedgewise.com/).\n",
     "\n",
     "Part of the Quantopian Lecture Series:\n",
     "\n",
     "* [www.quantopian.com/lectures](https://www.quantopian.com/lectures)\n",
     "* [github.com/quantopian/research_public](https://github.com/quantopian/research_public)\n",
     "\n",
     "\n",
     "---"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "##AR Models\n",
     "An autoregressive, or AR$(p)$, model is created by regressing a time series on its past values, its lags. The simplest form of an autoregressive model is an AR$(1)$ model, signifying using only one lag term. A first order autocorrelation model like this for a time series $x_t$ is:\n",
     "\n",
     "$$\n",
     "x_t = b_0 + b_1 x_{t - 1} + \\epsilon_t\n",
     "$$\n",
     "\n",
     "Where $x_{t - 1}$ represents the value of the time series at time $(t - 1)$ and $\\epsilon_t$ is the error term. We can extend this to an AR$(p)$ model, denoted:\n",
     "\n",
     "$$\n",
     "x_t = b_0 + b_1 x_{t-1} + b_2 x_{t - 2} \\ldots + b_p x_{t - p} + \\epsilon_t\n",
     "$$\n",
     "\n",
     "For an AR model to function properly, we must require that the time series is covariance stationary. This means that it follows three conditions:\n",
     "\n",
     "1. The expected value of the time series is constant and finite at all times, i.e. $E[y_t] = \\mu$ and $\\mu < \\infty$ for all values of $t$.\n",
     "2. The variance of the time series is constant and finite for all time periods.\n",
     "3. The covariance of the time series with itself for a fixed number of periods in either the future or the past is constant and finite for all time periods, i.e\n",
     "\n",
     "$$\n",
     "COV(y_t, y_{t - s}) = \\lambda, \\  |\\lambda| < \\infty, \\text{ $\\lambda$ constant}, \\  t = 1, 2, \\ \\ldots, T; \\  s = 0, \\pm 1, \\pm 2, \\ldots, \\pm T\n",
     "$$\n",
     "\n",
     "Note that this mathematical representation includes condition 2.\n",
     "\n",
     "If these conditions are not satisfied, our estimation results will not have real-world meaning. Our estimates for the parameters will be biased, making any tests that we try to form using the model invalid. Unfortunately, it can be a real pain to find a covariance-stationary time series in the wild in financial markets. For example, when we look at the stock price of Apple, we can clearly see an upward trend. The mean is increasing with time. There are ways, however to make a non-stationary time series stationary. Once we have performed this transformation, we can build an autoregressive models under the above assumptions."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "##Simulating Data\n",
     "\n",
     "Here we will draw data samples from a simulated AR$(3)$ process."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 1,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0sAAAHxCAYAAAC8iidMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
       "text/plain": [
        "<matplotlib.figure.Figure at 0x7fd3cfcc05d0>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "import numpy as np\n",
     "import pandas as pd\n",
     "from scipy import stats\n",
     "import statsmodels.api as sm\n",
     "import statsmodels.tsa as tsa\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
     "# ensures experiment runs the same every time\n",
     "np.random.seed(100)\n",
     "\n",
     "# This function simluates an AR process, generating a new value based on historial values,\n",
     "# autoregressive coefficients b1 ... bk, and some randomness.\n",
     "def AR(b, X, mu, sigma):\n",
     "    l = min(len(b)-1, len(X))\n",
     "    b0 = b[0]\n",
     "    \n",
     "    return b0 + np.dot(b[1:l+1], X[-l:]) + np.random.normal(mu, sigma)\n",
     "\n",
     "b = np.array([0, 0.8, 0.1, 0.05])\n",
     "X = np.array([1])\n",
     "\n",
     "mu = 0\n",
     "sigma = 1\n",
     "\n",
     "for i in range(10000):\n",
     "    X = np.append(X, AR(b, X, mu, sigma))\n",
     "    \n",
     "plt.plot(X)\n",
     "plt.xlabel('Time')\n",
     "plt.ylabel('AR Series Value');"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Note how this process fluctuates around some central value. This value is the mean of our time series. As we have a constant mean throughout time and the fluctuations seem to all stray within a given distance from the mean, we might hypothesize that this series is stationary. We would want to rigorously test that in practice, which we will explore lightly in the examples at the end of this lecture. Also see the [stationarity lecture](https://www.quantopian.com/lectures/integration-cointegration-and-stationarity) from the Quantopian Lecture Series.\n",
     "\n",
     "In this case, however, we have constructed the model to be stationary, so no need to worry about testing for stationarity right now."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "##Tail Risk\n",
     "\n",
     "Autoregressive processes will tend to have more extreme values than data drawn from say a normal distribution. This is because the value at each time point is influenced by recent values. If the series randomly jumps up, it is more likely to stay up than a non-autoregressive series. This is known as 'fat-tailledness' (fat-tailed distribution) because the extremes on the pdf will be fatter than in a normal distribution.\n",
     "\n",
     "Much talk of tail risk in finance comes from the fact that tail events do occur and are hard to model due to their infrequent occurrence. If we have reason to suspect that a process is autoregressive, we should expect risk from extreme tail events and adjust accordingly.\n",
     "\n",
     "AR models are just one of the sources of tail risk, so don't assume that because a series is non-AR, it does not have tail risk.\n",
     "\n",
     "We'll check for that behavior now."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 2,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "Frequency of std events in X \n",
       "1: 0.327567243276\t2: 0.185481451855\t3: 0.0909909009099\t4: 0.035796420358\n",
       "Frequency of std events in a normal process \n",
       "1: 0.158655253931\t2: 0.0227501319482\t3: 0.00134989803163\t4: 3.16712418331e-05\n"
      ]
     }
    ],
    "source": [
     "def compare_tails_to_normal(X):\n",
     "    # Define matrix to store comparisons\n",
     "    A = np.zeros((2,4))    \n",
     "    for k in range(4):             \n",
     "        #stores tail probabilities of the sample series vs a normal series\n",
     "        A[0, k] = len(X[X > (k + 1)]) / float(len(X)) # Estimate tails of X        \n",
     "        A[1, k] = 1 - stats.norm.cdf(k + 1) # Compare to Gaussian distribution\n",
     "    print 'Frequency of std events in X \\n1: %s\\t2: %s\\t3: %s\\t4: %s' % tuple(A[0])\n",
     "    print 'Frequency of std events in a normal process \\n1: %s\\t2: %s\\t3: %s\\t4: %s' % tuple(A[1])\n",
     "    return A\n",
     "\n",
     "compare_tails_to_normal(X);"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "##Estimations of Variance Will be Wrong\n",
     "\n",
     "Because an AR process has a tail heavy and non-normal distribution of outcomes, estimates of variance on AR processes will be wrong. This is dangerous because variance is used to calculate many quantities in staistics, most importantly confidence intervals and p-values. Because the width of the confidence interval is often based on a variance estimate, we can no longer trust p-values that come from AR processes. For more information on p-values please see the Hypothesis Testing notebook in the [Quantopian Lecture Series](https://www.quantopian.com/lectures).\n",
     "\n",
     "Let's check this here. First we'll define some helper functions that compute a naive 95% confidence interval for the true value of the mean on some input series."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 3,
    "metadata": {
     "collapsed": false
    },
    "outputs": [],
    "source": [
     "def compute_unadjusted_interval(X):\n",
     "    T = len(X)\n",
     "    # Compute mu and sigma MLE\n",
     "    mu = np.mean(X)\n",
     "    sigma = np.std(X)\n",
     "    # Compute the bounds using standard error\n",
     "    lower = mu - 1.96 * (sigma/np.sqrt(T))\n",
     "    upper = mu + 1.96 * (sigma/np.sqrt(T))\n",
     "    return lower, upper\n",
     "\n",
     "# We'll make a function that returns true when the computed bounds contain 0\n",
     "def check_unadjusted_coverage(X):\n",
     "    l, u = compute_unadjusted_interval(X)\n",
     "    # Check to make sure l <= 0 <= u\n",
     "    if l <= 0 and u >= 0:\n",
     "        return True\n",
     "    else:\n",
     "        return False\n",
     "    \n",
     "def simululate_AR_process(b, T):\n",
     "    X = np.array([1])\n",
     "\n",
     "    mu = 0\n",
     "    sigma = 1\n",
     "\n",
     "    for i in range(T):\n",
     "        X = np.append(X, AR(b, X, mu, sigma))\n",
     "        \n",
     "    return X"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Now we'll run an experiment 1000 times in which we compute an AR series, then estimate the mean and take a naive 95% confidence interval around it. Then we'll check if the confidence interval contains 0, the true long-term mean of our series, and record that in our `outcomes` array."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 4,
    "metadata": {
     "collapsed": false
    },
    "outputs": [],
    "source": [
     "trials = 1000\n",
     "outcomes = np.zeros((trials, 1))\n",
     "\n",
     "for i in range(trials):\n",
     "    #note these are the same values we used to generate the initial AR array\n",
     "    Z = simululate_AR_process(np.array([0, 0.8, 0.1, 0.05]), 100)\n",
     "    if check_unadjusted_coverage(Z):\n",
     "        # The internal contains 0, the true value\n",
     "        outcomes[i] = 1\n",
     "    else:\n",
     "        outcomes[i] = 0"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Finally let's check, if our test is calibrated correctly, then we should have the confidence interval contain 0, 95% of the time."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 5,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "text/plain": [
        "0.193"
       ]
      },
      "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
     "np.sum(outcomes) / trials"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Looks like something is severly wrong. What's going on here is that the AR series moves around a lot more, but the estimate of variance assumes stationarity and doesn't take into account all of that motion. As a result the confidence intervals are way smaller than they should be and don't contain the correct value nearly enough.\n",
     "\n",
     "This gives us a false sense of security. Stationarity tests should usually catch AR behavior and let us know that estimates of variance will be wrong. For more information please see the Integration, Cointegration, and Stationarity lecture of the [Quantopian Lecture Series](https://www.quantopian.com/lectures).\n",
     "\n",
     "####Correcting for Variance\n",
     "\n",
     "In practice it can be very difficult to accurately estimate variance on an AR series, but one attempt to do this is the Newey-West estimation. You can find information on it [here](https://en.wikipedia.org/wiki/Newey%E2%80%93West_estimator)."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "##Testing for AR Behavior\n",
     "\n",
     "In order to determine the order, $p$, of an AR$(p)$ model, we look at the autocorrelations of the time series. These are the correlations of the series with its past values. The $k$-th order autocorrelation is\n",
     "\n",
     "$$\n",
     "\\rho_k = \\frac{COV(x_t, x_{t - k})}{\\sigma_x^2} = \\frac{E[(x_t - \\mu)(x_{t - k} - \\mu)}{\\sigma_x^2}\n",
     "$$\n",
     "\n",
     "Where $k$ represents the number of periods lagged. We cannot directly observe the autocorrelations so we estimate them as \n",
     "\n",
     "$$\n",
     "\\hat{\\rho}_k = \\frac{\\sum_{t = k + 1}^T[(x_t - \\bar{x})(x_{t - k} - \\bar{x})]}{\\sum_{t = 1}^T (x_t - \\bar{x})^2}\n",
     "$$\n",
     "\n",
     "For our purposes, we can use a pair of tools called the autocorrelation function (ACF) and the partial autocorrelation function (PACF) in order to determine the order of our model. The PACF controls for shorter lags, unlike the ACF. These functions are included with many statistical packages and compute the sample autocorrelations for us, allowing us to determine the appropriate value of $p$.\n",
     "\n",
     "We will demonstrate these functions on our above example of a stationary series:"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 6,
    "metadata": {
     "collapsed": true
    },
    "outputs": [],
    "source": [
     "from statsmodels.tsa.stattools import acf, pacf"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 7,
    "metadata": {
     "collapsed": true
    },
    "outputs": [],
    "source": [
     "X = simululate_AR_process(np.array([0, 0.8, 0.1, 0.05]), 1000)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 8,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "Autocorrelations:\n",
       "[ 1.          0.57070553  0.58001725  0.88553976  0.55669882  0.57640436\n",
       "  0.78307405  0.54759415  0.55254723  0.70158722  0.53082572  0.52673077\n",
       "  0.62850967  0.50091633  0.48541475  0.56544198  0.46391848  0.45390828\n",
       "  0.51491108  0.43375521  0.43938017  0.46783793  0.40843913  0.42270738\n",
       "  0.42566802  0.38298857  0.38379558  0.40029717  0.36548704  0.36088258\n",
       "  0.37743521  0.34745858  0.34065687  0.35239327  0.33194099  0.31758452\n",
       "  0.33113111  0.31688646  0.2818213   0.31237461  0.31252031]\n",
       "\n",
       "Partial Autocorrelations:\n",
       "[ 1.          0.57127624  0.37827414  0.80672256  0.00162478  0.00490887\n",
       " -0.04497956  0.04013524 -0.05301909  0.00706455 -0.00528104  0.0051476\n",
       " -0.03421094 -0.03637203 -0.06841025  0.00396849 -0.02433695  0.03906542\n",
       "  0.02344142  0.02958713  0.07306916 -0.01201434  0.00279546  0.01328444\n",
       " -0.00679875 -0.01547705 -0.09186641  0.05847768  0.03298422  0.04395628\n",
       " -0.01605778  0.00286143 -0.00734411 -0.01496441  0.01152254 -0.01185183\n",
       "  0.01895763  0.01901218 -0.09532658  0.00791091  0.06173332]\n"
      ]
     }
    ],
    "source": [
     "# We'll choose 40 lags. This is a bit arbitrary, but you want to include all the lags you think might\n",
     "# feasibly impact the current value.\n",
     "nlags = 40\n",
     "# Note, this will produce nlags + 1 values, as we include the autocorrelation of\n",
     "# X[-1] with X[-1], which is trivially 1.\n",
     "# The reason this is done is because that is the 0th spot in the array and corresponds\n",
     "# to the 0th lag of X[(-1)-0].\n",
     "X_acf = acf(X, nlags=nlags)\n",
     "print 'Autocorrelations:\\n' + str(X_acf) + '\\n'\n",
     "X_pacf = pacf(X, nlags=nlags)\n",
     "print 'Partial Autocorrelations:\\n' + str(X_pacf)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Let's plot out the values now."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 9,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0UAAAH6CAYAAADIhOc8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAACUmlAEAACU2v8HIIapZ7f1SHYAAAAASUVORK5CYII=\n",
       "text/plain": [
        "<matplotlib.figure.Figure at 0x7fd3cfca3390>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "plt.plot(X_acf, 'ro')\n",
     "plt.xlabel('Lag')\n",
     "plt.ylabel('Autocorrelation')\n",
     "plt.title(\"ACF\");"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 10,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0sAAAH6CAYAAADWTdePAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAgAAKCAsAQAAFBCWAAAACghLAAAABf4/mCxAoaM3kMIAAAAASUVORK5CYII=\n",
       "text/plain": [
        "<matplotlib.figure.Figure at 0x7fd3cc2c5a50>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "plt.plot(X_pacf, 'ro')\n",
     "plt.xlabel('Lag')\n",
     "plt.ylabel('Autocorrelation')\n",
     "plt.title(\"PACF\");"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "####Statistical Testing\n",
     "\n",
     "Just looking at the graphs alone isn't enough. We need to use some degree of statistical rigor. The `acf` and `pacf` functions will return confidence intervals on all the autocorrelations. We can check if these intervals overlap with zero. If they do then we say that zero is within the set confidence interval for the true parameter value, and don't treat the lag as having any meaningful autocorrelation.\n",
     "\n",
     "NOTE: This only works if the assumptions underlying the confidence interval computations are satisfied. Please check these assumptions before you assume the test is meaningful. The assumptions will differ in every case, so please read the statistical documentation of your own test and go from there."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 11,
    "metadata": {
     "collapsed": true
    },
    "outputs": [],
    "source": [
     "# We have to set a confidence level for our intervals, we choose the standard of 95%,\n",
     "# corresponding with an alpha of 0.05.\n",
     "X_acf, X_acf_confs = acf(X, nlags=nlags, alpha=0.05)\n",
     "X_pacf, X_pacf_confs = pacf(X, nlags=nlags, alpha=0.05)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "After getting the confidence interval data, we'll write a function to plot it."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 12,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0UAAAH6CAYAAADIhOc8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQKEJRQAAQKEJRQAAQKEJRQAAQKH9f2IvuFNtPKHdAAAAAElFTkSuQmCC\n",
       "text/plain": [
        "<matplotlib.figure.Figure at 0x7fd3cc304d50>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "def plot_acf(X_acf, X_acf_confs, title='ACF'):\n",
     "    # The confidence intervals are returned by the functions as (lower, upper)\n",
     "    # The plotting function needs them in the form (x-lower, upper-x)\n",
     "    errorbars = np.ndarray((2, len(X_acf)))\n",
     "    errorbars[0, :] = X_acf - X_acf_confs[:,0]\n",
     "    errorbars[1, :] = X_acf_confs[:,1] - X_acf\n",
     "\n",
     "    plt.plot(X_acf, 'ro')\n",
     "    plt.errorbar(range(len(X_acf)), X_acf, yerr=errorbars, fmt='none', ecolor='gray', capthick=2)\n",
     "    plt.xlabel('Lag')\n",
     "    plt.ylabel('Autocorrelation')\n",
     "    plt.title(title);\n",
     "plot_acf(X_acf, X_acf_confs)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 13,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0sAAAH6CAYAAADWTdePAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCEsAAAAFhCUAAIAC/x+g19uGWcVyuAAAAABJRU5ErkJggg==\n",
       "text/plain": [
        "<matplotlib.figure.Figure at 0x7fd3cc287f50>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "plot_acf(X_pacf, X_pacf_confs, title='PACF')"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Notice how for the PACF, only the first three lags are the only ones that appear to be significantly different from $0$, which makes sense because we directly constructed an AR model of order $3$. However, these results may vary for each random series generated in this notebook.\n",
     "\n",
     "\n",
     "In a real-world time series, we use these plots to determine the order of our model. We would then attempt to fit a model using a maximum likelihood function."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "##Fitting a Model"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "We'll use one of the functions already implemented in Python to fit an AR model. We'll try this on our simulated data first."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 14,
    "metadata": {
     "collapsed": false
    },
    "outputs": [],
    "source": [
     "# Construct an unfitted model\n",
     "model = tsa.api.AR(X)\n",
     "# Fit it\n",
     "model = model.fit()"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "The `model` object has a lot of useful information on it, use the `?` notation to find out more. We'll be focusing on a few attributes, starting with `model.params` the estimated parameters in the model, one for each lag, and `model.bse`, the estimated standard error for each of the parameters."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 15,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "Parameters\n",
       "[ 0.01001762  0.04631952  0.06669125  0.84444655 -0.02662476  0.07031018\n",
       " -0.03883539  0.05362627 -0.06257934  0.03183551  0.02255835  0.06366225\n",
       " -0.02663838 -0.00824025 -0.09032174 -0.02259851 -0.05894006 -0.03575162\n",
       "  0.01899755  0.0265497   0.08158745 -0.00494997]\n",
       "Standard Error\n",
       "[ 0.03210283  0.03228516  0.03221285  0.03226938  0.04224386  0.04226218\n",
       "  0.04228528  0.04223737  0.04222499  0.04230633  0.04232241  0.04227582\n",
       "  0.04230175  0.04232492  0.04221253  0.04221065  0.04218858  0.04214434\n",
       "  0.04212185  0.03217742  0.03220652  0.03227438]\n"
      ]
     }
    ],
    "source": [
     "print 'Parameters'\n",
     "print model.params\n",
     "print 'Standard Error'\n",
     "print model.bse"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 16,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0sAAAH6CAYAAADWTdePAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCEsAAAAFhCUAAIACwhIAAECB/w+HJHl5qKYKygAAAABJRU5ErkJggg==\n",
       "text/plain": [
        "<matplotlib.figure.Figure at 0x7fd3cc2a5510>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
     "# To plot this we'll need to format a confidence interval 2D array like the previous functions returned\n",
     "# Here is some quick code to do that\n",
     "model_confs = np.asarray((model.params - model.bse, model.params + model.bse)).T\n",
     "\n",
     "plot_acf(model.params, model_confs, title='Model Estimated Parameters')"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "### Choosing the Number of Lags\n",
     "\n",
     "#### Estimations Will Yield Too Many Lags\n",
     "\n",
     "We can see our model estimated quite a few parameters. In this case we know there are too many because we simulated the data as an AR(3) process. The reason that AR models will estimate many more lags than is actually the case is due to indirect dependency. If $X_t$ depends on $X_{t-1}$, then indirectly and to a lesser extent it will depend on $X_{t-2}$. In the presence of more than one lag in the data generating process, we will get potentially complex harmonic structures in the lags. These indirect dependencies will be picked up by a simple estimation.\n",
     "\n",
     "#### You Want the Fewest Parameters That Yield a Decent Model\n",
     "\n",
     "In general it's rarely the case that you can get anything useful out of a model with many parameters, see the Overfitting lecture for why in the Quantopian Lecture Series. In this case we want to select a number of lags that we believe explains what is happening, but without overfitting and choosing a model with way too many lags. \n",
     "\n",
     "Observing the ACF and PACF indicates that only the first 3 lags may be useful. However, we will expand the number of lags to 10 to double-check our initial data. We will use information criterion, specifically Akaike Information Criterion (AIC) and Bayes Information Criterion (BIC) to decide the correct number of parameters. For more information on choosing models using information criterion, please see the corresponding lecture in the Quantopian Lecture Series.\n",
     "\n",
     "Interpreting the AIC and BIC is done as follows. Compute the AIC and BIC for all models we wish to consider, and note the smallest AIC and BIC recorded $AIC_{min}$ and $BIC_{min}$. These are the models which minimize information loss under each metric. For each type of IC We then can compute the *relative likelihood* of each model $i$ by taking \n",
     "\n",
     "$$l = e^{(IC_{min} - IC_{i})/2}$$\n",
     "\n",
     "We can interpret $l$ as model $i$ is $l$ times as likely to minimize information loss, compared to the minimum AIC model. It might take a few reads to understand this, so let's just see it in action."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 17,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "Relative Likelihoods\n",
       "[[ 0.54739169]\n",
       " [ 0.59052713]\n",
       " [ 0.99861675]\n",
       " [ 0.997749  ]\n",
       " [ 0.99628437]\n",
       " [ 0.99619915]\n",
       " [ 0.9960884 ]\n",
       " [ 1.        ]\n",
       " [ 0.99850924]\n",
       " [ 0.99809322]]\n",
       "Number of parameters in minimum AIC model 8\n"
      ]
     }
    ],
    "source": [
     "N = 10\n",
     "AIC = np.zeros((N, 1))\n",
     "\n",
     "for i in range(N):\n",
     "    model = tsa.api.AR(X)\n",
     "    model = model.fit(maxlag=(i+1))\n",
     "    AIC[i] = model.aic\n",
     "    \n",
     "AIC_min = np.min(AIC)\n",
     "model_min = np.argmin(AIC)\n",
     "\n",
     "print 'Relative Likelihoods'\n",
     "print np.exp((AIC_min-AIC) / 2)\n",
     "print 'Number of parameters in minimum AIC model %s' % (model_min+1)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 18,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "Relative Likelihoods\n",
       "[[ 0.5508575 ]\n",
       " [ 0.59280494]\n",
       " [ 1.        ]\n",
       " [ 0.99666667]\n",
       " [ 0.99274501]\n",
       " [ 0.99020386]\n",
       " [ 0.98763995]\n",
       " [ 0.98905709]\n",
       " [ 0.9851272 ]\n",
       " [ 0.98226451]]\n",
       "Number of parameters in minimum BIC model 3\n"
      ]
     }
    ],
    "source": [
     "N = 10\n",
     "BIC = np.zeros((N, 1))\n",
     "\n",
     "for i in range(N):\n",
     "    model = tsa.api.AR(X)\n",
     "    model = model.fit(maxlag=(i+1))\n",
     "    BIC[i] = model.bic\n",
     "    \n",
     "BIC_min = np.min(BIC)\n",
     "model_min = np.argmin(BIC)\n",
     "\n",
     "print 'Relative Likelihoods'\n",
     "print np.exp((BIC_min-BIC) / 2)\n",
     "print 'Number of parameters in minimum BIC model %s' % (model_min+1)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "Our conclusion is that the AIC estimates the 4 parameter model as most likely, whereas the BIC estimates 3. Because we are always looking for reasons to knock off a parameter, we choose the 3. In this case it happened to be the exact right answer, but this will not always be the case, especially in noisy real data. Don't assume that using this method will always get you the right answer."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "### Evaluating Residuals\n",
     "\n",
     "One final step we might do before performing an out of sample test for this model would be to evaluate its residual behavior. The AIC and BIC already do this to an extent, effectively measuring how much information is left on the table (in the residuals) after the model has made its predictions. For more information on residuals analysis see the Violations of Regression Models lecture. Here we'll just check for normality of the residuals."
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 19,
    "metadata": {
     "collapsed": false
    },
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
       "The residuals seem normally distributed.\n"
      ]
     }
    ],
    "source": [
     "model = tsa.api.AR(X)\n",
     "model = model.fit(maxlag=3)\n",
     "\n",
     "from statsmodels.stats.stattools import jarque_bera\n",
     "\n",
     "score, pvalue, _, _ = jarque_bera(model.resid)\n",
     "\n",
     "if pvalue < 0.10:\n",
     "    print 'We have reason to suspect the residuals are not normally distributed.'\n",
     "else:\n",
     "    print 'The residuals seem normally distributed.'"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "The residuals seem normally distributed. There are more model validation steps that could be done, but these are the core ones for an AR model. The next steps would be testing the model out of sample, and then using it to make predictions on your data."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "## References\n",
     "* \"Quantitative Investment Analysis\", by DeFusco, McLeavey, Pinto, and Runkle\n",
     "* \"Analysis of Financial Time Series\", by Ruey Tsay"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "*This presentation is for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation for any security; nor does it constitute an offer to provide investment advisory or other services by Quantopian, Inc. (\"Quantopian\"). Nothing contained herein constitutes investment advice or offers any opinion with respect to the suitability of any security, and any views expressed herein should not be taken as advice to buy, sell, or hold any security or as an endorsement of any security or company.  In preparing the information contained herein, Quantopian, Inc. has not taken into account the investment needs, objectives, and financial circumstances of any particular investor. Any views expressed and data illustrated herein were prepared based upon information, believed to be reliable, available to Quantopian, Inc. at the time of publication. Quantopian makes no guarantees as to their accuracy or completeness. All information is subject to change and may quickly become unreliable for various reasons, including changes in market conditions or economic circumstances.*"
    ]
   }
  ],
  "metadata": {
   "kernelspec": {
    "display_name": "Python 2",
    "language": "python",
    "name": "python2"
   },
   "language_info": {
    "codemirror_mode": {
     "name": "ipython",
     "version": 2
    },
    "file_extension": ".py",
    "mimetype": "text/x-python",
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython2",
    "version": "2.7.12"
   }
  },
  "nbformat": 4,
  "nbformat_minor": 0
 }