ARIMAX and seasonal splines

try_arima_exog.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:af177f371cc5a1aa3de78af1392a7906ee41a95d1a10256c34dc1d6da848edd1"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "ARMA, ARIMA with exogenous variables - seasonal pattern"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# -*- coding: utf-8 -*-\n",
      "\"\"\"\n",
      "Created on Wed Jun 24 10:24:44 2015\n",
      "\n",
      "Author: Josef Perktold\n",
      "License: \n",
      "\"\"\""
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 1,
       "text": [
        "'\\nCreated on Wed Jun 24 10:24:44 2015\\n\\nAuthor: Josef Perktold\\nLicense: \\n'"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The following illustrates the use of OLS, ARMA and ARIMA for use with splines to model the seasonal pattern.\n",
      "We use the elnino dataset that contains monthly temperatures and use spline basis functions created with `patsy` as explanatory variable, `exog`.\n",
      "\n",
      "I'm working with plain numpy arrays except for creating the spline basis with patsy. Specifically, I'm not familiar enough with date handling in `pandas` and `statsmodels.tsa` to write the example in a quick way."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "First, we import the required packages and load the data. We also create a `months` variable that encodes the month of the observation as an integer."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "import matplotlib.pyplot as plt\n",
      "import patsy\n",
      "import statsmodels.api as sm\n",
      "\n",
      "from statsmodels.datasets.elnino import load\n",
      "\n",
      "data = load()\n",
      "data.raw_data = data.raw_data[:-1]\n",
      "temperature = data.raw_data[:,1:].ravel()\n",
      "nobs = len(temperature)\n",
      "years_unique = data.raw_data[:,0]\n",
      "years = np.repeat(years_unique, 12)\n",
      "months = np.tile(np.arange(12), len(years_unique))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "months[:36]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11,  0,  1,  2,  3,  4,\n",
        "        5,  6,  7,  8,  9, 10, 11,  0,  1,  2,  3,  4,  5,  6,  7,  8,  9,\n",
        "       10, 11])"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "`Patsy` can create bspline basis function based on our month integer array, Where the degrees if freedom parameter `df` specifies the number of knots to use. The created design matrix includes a constant, since I didn't remove it explicitly in the formula. We will need to include the constant in the OLS model, but we will need to drop it for the ARMA and ARIMA models that include a constant directly. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "exog_season = patsy.dmatrix('bs(months, df=4)')\n",
      "print(exog_season.shape)\n",
      "print(exog_season[:5])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "(720, 5)\n",
        "[[ 1.          0.          0.          0.          0.        ]\n",
        " [ 1.          0.40721262  0.04357626  0.00150263  0.        ]\n",
        " [ 1.          0.58001503  0.15026296  0.01202104  0.        ]\n",
        " [ 1.          0.58151766  0.28399699  0.040571    0.        ]\n",
        " [ 1.          0.47483095  0.40871525  0.09616829  0.        ]]\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "res_ols = sm.OLS(temperature, exog_season).fit()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We will predict the time series for new observations at the end of our sample. `horizon` specifies the number of months to forecast. Since the `exog` are forcasting the seasonal pattern, we can reuse the array of explanatory variables from the start of the series. In general, we would have to get the correct months for extending the sample to the forecast period."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "horizon = 36\n",
      "n_disp = 6 * 12\n",
      "predicted_ols = np.concatenate((res_ols.fittedvalues[-n_disp:], \n",
      "                                res_ols.predict(exog_season[:horizon])))\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Next, we estimate the ARMAX(1,1) and the ARIMAX(1,1,1) models and use them for prediction"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# the next two are to try out the ARMA and ARIMA model for 1-d exog\n",
      "# exog_season2_ = np.column_stack((np.ones(nobs), (months - months.mean())**2))\n",
      "# exog_season2 = exog_season2_[:, 1]    # bug only 1d supported, works now ?\n",
      "\n",
      "exog_season2 = exog_season[:, 1:]  # bug/requirement: no constant in exog\n",
      "\n",
      "res_arma = sm.tsa.ARMA(temperature, order=(3,1), exog=exog_season2).fit(iprint=0)\n",
      "\n",
      "predicted_arma = res_arma.predict(start=nobs, end=nobs+horizon-1, exog=exog_season2[:horizon])    "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# With ARIMA, we use differencing only to have a test case. \n",
      "# I don't think it makes much sense for the temperature\n",
      "\n",
      "res_arima = sm.tsa.ARIMA(temperature, order=(3,1,1), exog=exog_season2).fit(iprint=0, \n",
      "                                                    start_params=res_arma.params, \n",
      "                                                    solver='nm', maxiter=5000, maxfun=5000)\n",
      "\n",
      "res_arima_fittedvalues = res_arima.predict(typ='levels')\n",
      "predicted_arima = res_arima.predict(start=nobs, end=nobs+horizon-1, exog=exog_season2[:horizon], typ='levels')    \n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Optimization terminated successfully.\n",
        "         Current function value: 0.699932\n",
        "         Iterations: 3309\n",
        "         Function evaluations: 4761\n"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "And finally, we plot the data and our three predictions. The insample predictions are the one-step ahead prediction from fitting the model.\n",
      "\n",
      "I would like to make just a few comments on the results.\n",
      "\n",
      "We estimated a seasonal pattern that is deterministic and constant over time. The prediction of the seasonal component does not depend on the seasonal component in the previous year. This is in contrast to a seasonal ARIMA where we obtain the seasonal prediction from the seasonal differencing and effects of seasonal lags. A SARIMAX model would predict that a cold year is followed by another cold year, while the deterministic pattern predicts an *average* season defined by our estimated spline.\n",
      "\n",
      "The parameter estimates for the spline basis functions are almost the same for OLS and ARMA. As a consequence they produce the same long run prediction. The prediction of ARMA differs form the one of OLS only in the short run, where ARMA adjusts the prediction to the previous prediction errors.\n",
      "\n",
      "Both estimated parameters and prediction of the ARIMA model differ from OLS and ARMA because the estimation is done for monthly differences"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fig = plt.figure(figsize=(12, 8))\n",
      "ax = fig.add_subplot(1,1,1)\n",
      "\n",
      "ax.plot(temperature[-n_disp:], '--', label='data')\n",
      "ax.plot(temperature[-n_disp:], 'o', alpha=0.5)\n",
      "\n",
      "ax.plot(predicted_ols, lw=3, alpha=0.5, label='OLS')\n",
      "\n",
      "ax.plot(np.concatenate((res_arma.fittedvalues[-n_disp:], predicted_arma)), lw=2, alpha=0.5, label='ARMA')\n",
      "\n",
      "ax.plot(np.concatenate((res_arima_fittedvalues[-n_disp:], predicted_arima)), lw=2, alpha=0.5, label='ARIMA')\n",
      "\n",
      "\n",
      "predicted_arma2 = res_arma.predict(start=nobs, end=nobs+horizon - 1, exog=exog_season2[:horizon + 11])   \n",
      "#ax.plot(np.concatenate((res_arma.fittedvalues[-n_disp:], predicted_arma2)), lw=2, alpha=0.25, label='ARMA2')\n",
      "\n",
      "predicted_arima2 = res_arima.predict(start=nobs, end=None, exog=exog_season2[12:horizon+12], typ='levels')    \n",
      "#ax.plot(np.concatenate((res_arima_fittedvalues[-n_disp:], predicted_arima2)), lw=2, alpha=0.5, label='ARIMA2')\n",
      "\n",
      "ax.legend(loc='upper left')\n",
      "print(np.max(np.abs(predicted_arma2 - predicted_arma)))\n",
      "print(predicted_arma2 - predicted_arma)\n",
      "print(len(predicted_ols) - n_disp, len(predicted_arma), len(predicted_arima), len(predicted_arma2))\n",
      "pred_ols = predicted_ols[-len(predicted_arma):]\n",
      "forecast = np.column_stack((pred_ols, predicted_arma, predicted_arima,\n",
      "                            predicted_arma - pred_ols, predicted_arima - pred_ols))\n",
      "import pandas as pd\n",
      "print(pd.DataFrame(forecast))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "0.0\n",
        "[ 0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.\n",
        "  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]\n",
        "36 36 36 36\n",
        "            0          1          2         3         4\n",
        "0   24.373629  24.851984  25.104862  0.478354  0.731233\n",
        "1   25.902156  26.301755  26.174447  0.399599  0.272291\n",
        "2   26.126282  26.453395  26.432472  0.327113  0.306190\n",
        "3   25.430381  25.715130  25.955835  0.284749  0.525454\n",
        "4   24.198829  24.448297  24.909398  0.249467  0.710569\n",
        "5   22.816001  23.041030  23.556861  0.225029  0.740860\n",
        "6   21.658644  21.856870  22.255483  0.198227  0.596839\n",
        "7   20.928099  21.098501  21.339532  0.170402  0.411433\n",
        "8   20.650300  20.790575  21.004493  0.140275  0.354192\n",
        "9   20.843555  20.954471  21.302082  0.110916  0.458527\n",
        "10  21.526171  21.609091  22.140268  0.082921  0.614098\n",
        "11  22.716454  22.774990  23.283266  0.058536  0.566812\n",
        "12  24.373629  24.470849  25.129737  0.097219  0.756108\n",
        "13  25.902156  25.958840  26.197938  0.056684  0.295783\n",
        "14  26.126282  26.163052  26.459417  0.036771  0.333135\n",
        "15  25.430381  25.461664  25.983399  0.031283  0.553018\n",
        "16  24.198829  24.232735  24.936793  0.033906  0.737964\n",
        "17  22.816001  22.854444  23.584184  0.038443  0.768184\n",
        "18  21.658644  21.697478  22.282808  0.038834  0.624165\n",
        "19  20.928099  20.961053  21.366863  0.032954  0.438764\n",
        "20  20.650300  20.672858  21.031825  0.022558  0.381524\n",
        "21  20.843555  20.853153  21.329414  0.009598  0.485859\n",
        "22  21.526171  21.522201  22.167600 -0.003970  0.641430\n",
        "23  22.716454  22.700278  23.310597 -0.016176  0.594143\n",
        "24  24.373629  24.406729  25.157069  0.033100  0.783440\n",
        "25  25.902156  25.903736  26.225270  0.001581  0.323114\n",
        "26  26.126282  26.115744  26.486749 -0.010538  0.360467\n",
        "27  25.430381  25.421018  26.010731 -0.009363  0.580350\n",
        "28  24.198829  24.197832  24.964125 -0.000997  0.765296\n",
        "29  22.816001  22.824461  23.611516  0.008461  0.795516\n",
        "30  21.658644  21.671728  22.310140  0.013085  0.651497\n",
        "31  20.928099  20.938935  21.394195  0.010836  0.466097\n",
        "32  20.650300  20.653863  21.059157  0.003562  0.408857\n",
        "33  20.843555  20.836837  21.356746 -0.006718  0.513191\n",
        "34  21.526171  21.508187  22.194932 -0.017983  0.668762\n",
        "35  22.716454  22.688242  23.337930 -0.028212  0.621476\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAAHfCAYAAABXrM5GAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0xe803160>"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "For completeness, we can look at the `summary` results of the three models. `x1` to `x4` are the spline components i.e. the four basis functions.\n",
      "\n",
      "As can be seen from the low Durbin Watson statistic in OLS, there remain a large amount of serial correlation when estimating with OLS. (The OLS parameter coefficients are still consistent, but the standard errors are incorrect because we don't use a `cov_type` that takes the unspecified serial correlation into account.)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print(res_ols.summary())"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "                            OLS Regression Results                            \n",
        "==============================================================================\n",
        "Dep. Variable:                      y   R-squared:                       0.764\n",
        "Model:                            OLS   Adj. R-squared:                  0.763\n",
        "Method:                 Least Squares   F-statistic:                     579.2\n",
        "Date:                Sun, 28 Jun 2015   Prob (F-statistic):          1.41e-222\n",
        "Time:                        13:37:46   Log-Likelihood:                -1081.1\n",
        "No. Observations:                 720   AIC:                             2172.\n",
        "Df Residuals:                     715   BIC:                             2195.\n",
        "Df Model:                           4                                         \n",
        "Covariance Type:            nonrobust                                         \n",
        "==============================================================================\n",
        "                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
        "------------------------------------------------------------------------------\n",
        "const         24.3736      0.132    184.637      0.000        24.114    24.633\n",
        "x1             4.2329      0.298     14.181      0.000         3.647     4.819\n",
        "x2            -4.3298      0.305    -14.187      0.000        -4.929    -3.731\n",
        "x3            -4.3159      0.272    -15.877      0.000        -4.850    -3.782\n",
        "x4            -1.6572      0.183     -9.054      0.000        -2.017    -1.298\n",
        "==============================================================================\n",
        "Omnibus:                      148.044   Durbin-Watson:                   0.177\n",
        "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              305.297\n",
        "Skew:                           1.144   Prob(JB):                     5.08e-67\n",
        "Kurtosis:                       5.223   Cond. No.                         11.3\n",
        "==============================================================================\n",
        "\n",
        "Warnings:\n",
        "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print(res_arma.summary())"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "                              ARMA Model Results                              \n",
        "==============================================================================\n",
        "Dep. Variable:                      y   No. Observations:                  720\n",
        "Model:                     ARMA(3, 1)   Log Likelihood                -428.729\n",
        "Method:                       css-mle   S.D. of innovations              0.438\n",
        "Date:                Sun, 28 Jun 2015   AIC                            877.457\n",
        "Time:                        13:37:47   BIC                            923.250\n",
        "Sample:                             0   HQIC                           895.136\n",
        "                                                                              \n",
        "==============================================================================\n",
        "                 coef    std err          z      P>|z|      [95.0% Conf. Int.]\n",
        "------------------------------------------------------------------------------\n",
        "const         24.3944      0.165    147.615      0.000        24.071    24.718\n",
        "x1             4.1570      0.132     31.480      0.000         3.898     4.416\n",
        "x2            -4.3030      0.229    -18.778      0.000        -4.752    -3.854\n",
        "x3            -4.3513      0.148    -29.464      0.000        -4.641    -4.062\n",
        "x4            -1.7085      0.058    -29.352      0.000        -1.823    -1.594\n",
        "ar.L1.y        0.5084      0.441      1.153      0.249        -0.356     1.373\n",
        "ar.L2.y        0.4697      0.459      1.024      0.306        -0.430     1.369\n",
        "ar.L3.y       -0.1449      0.073     -1.994      0.047        -0.287    -0.002\n",
        "ma.L1.y        0.5575      0.443      1.257      0.209        -0.312     1.427\n",
        "                                    Roots                                    \n",
        "=============================================================================\n",
        "                 Real           Imaginary           Modulus         Frequency\n",
        "-----------------------------------------------------------------------------\n",
        "AR.1           -1.6084           +0.0000j            1.6084            0.5000\n",
        "AR.2            1.1643           +0.0000j            1.1643            0.0000\n",
        "AR.3            3.6865           +0.0000j            3.6865            0.0000\n",
        "MA.1           -1.7936           +0.0000j            1.7936            0.5000\n",
        "-----------------------------------------------------------------------------\n"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print(res_arima.summary())"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "                             ARIMA Model Results                              \n",
        "==============================================================================\n",
        "Dep. Variable:                    D.y   No. Observations:                  719\n",
        "Model:                 ARIMA(3, 1, 1)   Log Likelihood                -503.251\n",
        "Method:                       css-mle   S.D. of innovations              0.482\n",
        "Date:                Sun, 28 Jun 2015   AIC                           1026.502\n",
        "Time:                        13:37:47   BIC                           1072.280\n",
        "Sample:                             1   HQIC                          1044.176\n",
        "                                                                              \n",
        "==============================================================================\n",
        "                 coef    std err          z      P>|z|      [95.0% Conf. Int.]\n",
        "------------------------------------------------------------------------------\n",
        "const        186.9456     50.589      3.695      0.000        87.793   286.099\n",
        "x1            -1.3411      0.140     -9.577      0.000        -1.616    -1.067\n",
        "x2            -5.2950      0.147    -35.925      0.000        -5.584    -5.006\n",
        "x3            -0.9583      0.131     -7.325      0.000        -1.215    -0.702\n",
        "x4            -0.7035      0.080     -8.784      0.000        -0.860    -0.547\n",
        "ar.L1.D.y      1.1489        nan        nan        nan           nan       nan\n",
        "ar.L2.D.y     -0.2246        nan        nan        nan           nan       nan\n",
        "ar.L3.D.y      0.0756        nan        nan        nan           nan       nan\n",
        "ma.L1.D.y     -0.9978      0.001   -823.048      0.000        -1.000    -0.995\n",
        "                                    Roots                                    \n",
        "=============================================================================\n",
        "                 Real           Imaginary           Modulus         Frequency\n",
        "-----------------------------------------------------------------------------\n",
        "AR.1            1.0000           -0.0000j            1.0000           -0.0000\n",
        "AR.2            0.9847           -3.5004j            3.6362           -0.2064\n",
        "AR.3            0.9847           +3.5004j            3.6362            0.2064\n",
        "MA.1            1.0022           +0.0000j            1.0022            0.0000\n",
        "-----------------------------------------------------------------------------\n"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Correcting for autocorrelation in OLS**"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "res_ols2 = sm.OLS(temperature, exog_season).fit(cov_type='HAC', cov_kwds={'maxlags':4})\n",
      "print(res_ols2.summary())"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "                            OLS Regression Results                            \n",
        "==============================================================================\n",
        "Dep. Variable:                      y   R-squared:                       0.764\n",
        "Model:                            OLS   Adj. R-squared:                  0.763\n",
        "Method:                 Least Squares   F-statistic:                     278.6\n",
        "Date:                Sun, 28 Jun 2015   Prob (F-statistic):          2.97e-144\n",
        "Time:                        13:37:47   Log-Likelihood:                -1081.1\n",
        "No. Observations:                 720   AIC:                             2172.\n",
        "Df Residuals:                     715   BIC:                             2195.\n",
        "Df Model:                           4                                         \n",
        "Covariance Type:                  HAC                                         \n",
        "==============================================================================\n",
        "                 coef    std err          z      P>|z|      [95.0% Conf. Int.]\n",
        "------------------------------------------------------------------------------\n",
        "const         24.3736      0.118    207.421      0.000        24.143    24.604\n",
        "x1             4.2329      0.239     17.708      0.000         3.764     4.701\n",
        "x2            -4.3298      0.396    -10.920      0.000        -5.107    -3.553\n",
        "x3            -4.3159      0.234    -18.441      0.000        -4.775    -3.857\n",
        "x4            -1.6572      0.110    -15.105      0.000        -1.872    -1.442\n",
        "==============================================================================\n",
        "Omnibus:                      148.044   Durbin-Watson:                   0.177\n",
        "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              305.297\n",
        "Skew:                           1.144   Prob(JB):                     5.08e-67\n",
        "Kurtosis:                       5.223   Cond. No.                         11.3\n",
        "==============================================================================\n",
        "\n",
        "Warnings:\n",
        "[1] Standard Errors are heteroscedasticity and autocorrelation robust (HAC) using 4 lags and without small sample correction\n"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "res_arima.resid[-10:]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 14,
       "text": [
        "array([-0.33496999,  0.88996087,  0.00848455,  0.53244565,  0.23848789,\n",
        "       -0.15362877, -0.17228117, -0.11761784, -0.4848151 ,  0.13962137])"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 14
    }
   ],
   "metadata": {}
  }
 ]
}