Skip to content

Instantly share code, notes, and snippets.

@ChadFulton
Created November 18, 2013 21:06
Show Gist options
  • Save ChadFulton/7535278 to your computer and use it in GitHub Desktop.
Save ChadFulton/7535278 to your computer and use it in GitHub Desktop.
Statsmodels ARIMA Example
{
"metadata": {
"name": ""
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "code",
"collapsed": false,
"input": [
"import numpy as np\n",
"import pandas as pd\n",
"import statsmodels.api as sm\n",
"import matplotlib.pyplot as plt\n",
"\n",
"print 'Numpy:', np.__version__\n",
"print 'Pandas:', pd.__version__\n",
"import statsmodels\n",
"print 'Statsmodels:', statsmodels.__version__\n",
"\n",
"np.set_printoptions(precision=2)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Numpy: 1.8.0.dev-dff8c94\n",
"Pandas: 0.11.1.dev-06cd915\n",
"Statsmodels: 0.5.0.dev-4b58321\n"
]
}
],
"prompt_number": 59
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# In your example, you weren't able to predict because your dataframe was not sorted correctly;\n",
"# If you look at the model summary table, you'll see that the sample runs from\n",
"# 08-01-2013 to 05-01-2000\n",
"data = pd.read_csv('data/LocationIndexNew - Updated.csv', parse_dates='date', index_col=0, thousands=',').sort_index()\n",
"data['Unit Sales'].plot()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 94,
"text": [
"<matplotlib.axes.AxesSubplot at 0x109553cd0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEQCAYAAABV+ASvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsvXmcG/V9///SNZJG996H1zc+8GKDzWFCAAfaQI7SUEoS\nSPqAJP01TUIThy9tvykYQpymP5ISb9IEmiYhTtM2AZPmAAIBg9cc5rSx17e99q69h7Sn7tGtz/eP\n0WglraSZkTTsavbz/MsazX5Gb8/Me97z+rw/77eGEEJAoVAoFFWinesfQKFQKBTloE6eQqFQVAx1\n8hQKhaJiqJOnUCgUFUOdPIVCoagY6uQpFApFxejLfclxHL7zne9kP589exY///nP0dfXhyeffBIc\nx+HOO+9Ed3c3AMjeTqFQKBRlKevkWZbFAw88AAA4d+4cnn32WRBCsGvXLmzbtg3hcBg9PT3o7u5G\nOp2WvH3dunXQaDTviYEUCoWykCnr5HN59tln8aEPfQhutxvt7e1gGAYMw6C1tRVutxuEEMnbPR4P\n2tvblbSLQqFQKJDo5IPBIKamprBkyRKcOnUKLMti586dAPhon+M4pFIpWdspFAqFojySJl53796N\n66+/HgBgtVrBcRxuv/123HbbbYhGo+jq6pK9nUKhUCjKIxrJp1IpHDhwAA8++CAAoK2tDaOjo+A4\nDul0GoFAAAzDyN5eihdffLF21lEoFMoCwel0YtOmTbO2izr5t99+G5s2bYJWywf9Wq0Wt956K7Zv\n3w4AuOOOOyraXg7hrYFCoVAo0igZIJN5xu7du6seY8eOHTX4JeXZs2eP4segdkiH2iGd98IOQtRj\nS73YUcp30sVQFAqFomI0hMyvevIvvvgilWsoFApFJqV8J43kKRQKRcWo0sn39PQofoze3l7Fj0Ht\nkA61QzrvhR2AemypdztU6eQpFAqFwkM1eQqFQlEBVJOnUCiUBYgqnTzV6aRD7ZAOtUMearGl3u1Q\npZOnUCgUCg/V5CkUCkUFUE2eQqFQFiCqdPJUp5MOtUM61A55qMWWerdDlU6eQqFQKDxUk6dQKBQV\nQDV5CoVCWYCo0slTnU461A7pUDvkoRZb6t0OVTp5CoVCofBQTZ5CoVBUANXkKRQKZQGiSidPdTrp\nUDukQ+2Qh1psqXc7VOnkKRQKhcJDNXkKhUJRAVSTp1AolAWIKp081emkQ+2QDrVDHmqxpd7tUKWT\np1AoFAoP1eQpFApFBVBNnkKhUBYgqnTyVKeTDrVDOtQOeajFlnq3Q5VOnkKhUCg8opq83+/HI488\ngmg0igsuuACf/vSn0dfXhyeffBIcx+HOO+9Ed3c3AMjeXgyqyVMoFIp8SvlOvdgfPvbYY/jYxz6G\ntWvXAgDS6TR27dqFbdu2IRwOo6enB93d3bK2r1u3DhqNpvZWUigUCiWPsnJNOp3GiRMnsg4eADwe\nD9rb28EwDFwuF1pbW+F2u2Vt93g8ihpFdTrpUDukQ+2Qh1psqXc7ykbyoVAILMvikUceQTgcxo03\n3gij0QiWZbFz504AAMuy4DgOqVRK1nYKZS4YnA5jaYNlrn8GhfLeQUT41re+RcLhMIlEIuT+++8n\nIyMj5Ic//CGJxWIkGo2SRx99lMRiMdnbS7F7926xn0ShVEQilSJXfXcPmQhG5/qnUCg1p5TvFM2u\naWxshNfrhdFohE6nQ0tLC0ZHR8FxHMLhMAKBABiGQVtbm6zt5ch9Pert7aWf6eeafB71RxFLpvH8\ny/vmxe+hn+nnWn8uitjTYWJigmzbto088MADZO/evYQQQg4ePEjuvvtucvfdd5NDhw5l95W7Xc7T\nSA47duyoegwx9uzZo/gxqB3SkWJH7+lxcum3XyRvDk5VdIz5Yke1vBd2EKIeW+rFjlK+UzS7pqmp\nCd/4xjfytm3YsAEPP/zwrH3lbqdQcjnuCWBtm12x8QemwgAAXySh2DEolPkGrV1DmRdMhmL48KOv\nYe9XroWZ0SlyjAeeOYY9pydw1zUr8PGNixQ5BoUyV9DaNZR5zdmpMAj47BelGJgOY0OnA75IXLFj\nUCjzDVU6eZo7K535YocgpZydrMzJi9mRJgSDUxwuXuSAl6tMrllI56MWqMWWerdDlU6eUn8MTHFo\ntDA4O6VMJD8ejMHC6NDlYqkmT1lQUE2eMi/4/K8OYEWTBW5/FDtu2VDz8fcNTOG/3jqPOzcvwc/e\nGMSjn9hY82McHPZhkdOMJqux5mNTKGJQTZ4yrxmcCuP6VS2KRfKDU2Esa7TAaTbAV6FcI8bd/9uH\nv/zpG/jykwcRjNK3Bcr8QJVOnup00pkPdvgiCcSSaWxY5MBUOA4unpR9DDE7zk6FsayRhYtl4K1Q\nrilnRzKdBpdI4Q9fuArxZBpvn/dWdIz5cD5qhVpsqXc7VOnkKfWFEGXrtVosdrEYnK59baOBSQ7L\nGi1wmAzwRxIQVMqfvTGIYW/1xwtEkrAZ9WAZPTZ1uXDME6x6TAqlJlS9zKrG0No1C49fHxwmX//D\nUUIIIfc+dYQ8dXi0puOn02ly3ff3kqkQXzPp2p5eEojECSGE3Pwf+8je0xNVH+PsZIjc8pPXCSGE\nvHpmgnzx8QNVj0mhyKHi2jUUitIMTnFY3shXhlzRZMHZyTDiyTTue/ooJkKxqsePJFKIJdNwsQYA\n4HX5SALJVBqj/ijCFchDhfgjCThM/PhrWu047glm3xYolLlElU6e6nTSmQ92DEyFsTTj5Jc3WnB2\nKoxHXz2LF06M4dUzk5KOUc6OYCwJu0mfbVTjYhl4uQTcgShShICLp6q2wxdJwJl5iDRaGLCMDiO+\niKRxc5kP56NWqMWWerdDlU6eUl8MZDR5AFjWZMGBIR/+eNyDu65ZiTcHp6sePxhNwmo0ZD87zQZ4\nI3Gc9/JOOByrbSQPAGvb7Dg2RnV5ytyjSie/detWxY+xZcsWxY+xEOzg4kn4Igm0200AgE6HGQad\nBttuXIsb1rbinfNepNLiskc5O4IxflJUwMnyk6/DPn7CNZyQFsmXs8MfScBhnnHyF7bZcMwdkDRu\nLnN9PmqJWmypdztU6eQp9cOp8RCWN1mg0/JSik6rwTN/exWuXNaIFpsRDRYjTlQZEQejvFwj4DTz\ncs2QN4I2u7E2kXw0AYd55hhrW204TiN5yjxAlU6e6nTSmWs7jrgD6G7PLy9sMsxUobxiqQtvnROX\nbMpq8tEErDmRvCsz8Xrey2FNi60mmnwxuebkWBBpmZOvc30+aolabKl3O1Tp5Cn1w9EiTj6XK5Y0\nVK3L8xOvOZo8a4CXi2PYG8HqVltFi68KKZRrHGYDnGYD3h32VT02hVINtHYNZU656Uf78G+3Xowl\nDWzR77l4Eh965DU898X3V1xn/sf7BpBME3zh/csBAC/3T2LXu8M4MOTDt25ahyffHcG/3XpxxTYA\nwN/8cj8+//7l2NTlym779cER/PT1AViNBvzLTeuwosla1TEolHLQ2jWUecdUOI5QLIkul7nkPiyj\nx4pmC46NyZ/EFAhGCyZezQYcHwuiycrAaTLUJE/eF0nAmRPJA8AtF3fi6b+9CqtarDg04q/6GBRK\nJajSyVOdTjpzaccRtx/r2u3QZvLXS9FqM2EqVL7RR/k8+UTexKsrk13T5TTDYtQjrIAmL6DVaOBi\nDYhKzOBRy3UFqMeWerdDlU6eUh8ccwewroweL9DAMpjiKu/mFIom8yZehYi7y8WCZXRVa/KEEPij\nyTxNPhezQYeIRCdPodQaVTp5mjsrHSl2JFNp7BuYqvgYpewolllTjAaLAdPh8k6+nB2Bgjx5q1EP\nnVaDLhcLC6NHOFZdnnw4noJJr4VBV/x2Mul1iCbSko6hlusKUI8t9W6HKp08pbYcdgfw1V8fwqhf\n/jL9UqQJwTFPUHIkP11FJM/nyc9E2RqNBk6zAV0uMyyMDlw8VVWdGV9BZk0hJhrJU+YQVTp5qtNJ\nR4odp8dD0Gg02PXuSEXHKGbH+WkOdpMeLpYR/fsGCyMayefa4eXi+M83z2U/h2L5cg0AvH95I9a0\n2qDXaaHTahBLikfapc5HKT1ewGzQUk1eIebLPVItVJOnzCmnJ0K4bVMXnjriRkTiJKUYp8ZDWN1i\nk7Rvo0xNfmAqjF0Hh7OfA9EEbKZ8J3/fjWvRnGnTZzHqJE++FkNKJB+V8BChUJRAlU6e6nTSkWJH\n/0QI165swiWdDjxz1C37GMXsODMZwspmi6S/lxLJ59rBxVOYCMWRShOk0gSRRGpWJJ8La5A2+Zpr\nRyyZwo49pwHMXghViJyJV7VcV4B6bKl3O1Tp5Cm1I5UmODMZxspmKz65qQtPVCjZFNI/GcZKiYuD\nBE1eqm4ejieRShN4uTjC8SRYRl82TVNOGqXAVDiO/3lnCGPBKPzR2TnyuZgNupq9AVEoclGlk6c6\nnXTE7BjxR+BiDbAa9djY5YSXi2M8KK+RRzE7+idCWNEszcmbDDowOi2CZQqJ5dohZMuMBWMIFBQn\nK4aF0UlaEJVrh1Dv5sCQT1STN+m1iCapJq8E8+EeqQVUk6fMGf3jIazMOGONRoPuDjsOj1a3epOL\nJzEVjmORs/RK10IaJUg2AoLDHg/Gik66FiInjVJAmEjdf94rKteYDNJTKCmUWqNKJ091OumI2XFq\nYsbJA8D6Dgf6ZDr5QjvOTvJNQoTywlIQWxCVa0coE2WPh6IIRhN5OfLFkLogKtcOLpFCA8tg/5Bv\nVpnhQswM1eSVYj7cI7WAavKUOaN/IoQLCpz84SrrsPRPhrGiSdqkq4CUyVeBcCyJRguDsUCMbxhS\nRkoB+EhearlhgUg8hQvbbQjFkjg1HhKRa3SSUygplFoj6uR/+MMf4t5778WDDz6IvXv3AgD6+vpw\n//3345577sGRI0ey+8rdrhRUp5OOmB2nC5z8hW12nJ4MIS4jJbDQjv6JkOyKjGILovI0+XgSyxot\nGA/FZhUnKwZbgSYfSaRgYfh5inPTnEh2jVZyJK+W6wpQjy31bkf5qx+8DvvVr34VTU1NAIB0Oo1d\nu3Zh27ZtCIfD6OnpQXd3t6zt69atyzZVpsxfQrEkprl87dzM6LDExeLEWBDrOx0VjXtmMoyrljfK\n+ht5kXwKyxpZ9E+EZrX+K4aFkZ9dwyVSMBm0WN/hwEunJkSza2iePGWukCTX5KaueTwetLe3g2EY\nuFwutLa2wu12y9ru8XgUMwigOt2URGcIlLfj7GQYyxpma+cXdThkTb4W2sHnyMuL5MUWROXaIUTy\nY0FBrhFx8kZpkXyuHdFECqyBj+QBlI3k9ZmaNomUuKOfz9eVXNRiS73bIRrJm0wmfO9730NXVxdu\nueUWhEIhsCyLnTt3AgBYlgXHcUilUrK2U5Th5FgQX3j8Xbz05WuqHssTjKLdYZq1/aIOB17un6ho\nzKlwHOk0QZNFvJxBLnwkn5C0bziewrJGCyZCMQQiCSxtLN6QRICtQJPn4imYDVqsaLLgy9euhNlQ\nvqGJsCCqVBEzCkUpRK+4z372s/jmN7+Jq666Cr/5zW9gtVrBcRxuv/123HbbbYhGo+jq6pK9vRy5\nGlhvb6/sz3fddVdVfy/ls7BNqfEBXqeT+/c/fPZtBGPJ7OIbsf2FYxT7fiIYQ9w3Mev7yNBx9I36\nQQiRfT7+98XX0GhIZuU6qfYJmryU8zHhDaCBZcAyehw6O4Lhs/1lxz/XfzLbzFvq+YgkUjAzOuzd\nuxdd3FlRe0x6LaKJdFXno1af34v7I/ffStpz1113Kf7/lauXK2WPcIxqxysKkcipU6fIz3/+c5JK\npci9995LvF4vmZqaIg899BAhhMjeXordu3dL/Ukl2bFjR9VjiLFnzx7FjyHXDi6WJNd9fy/Z8r29\nZMjLSfqbcnZ896VT5D/fHJy1PZ1Ok6t39JJgNCHpGLl2PHfMTb72+8OS/i6XIS9H/uzfXyv5fa4d\nH37kVeIJRMjtO98kN/zwFfJy/0TZsV89M0nueuJd0d+Qa8dDL5wgv9p/XvyHZ7j5P/aRwamw6H7z\n8bqqFLXYUi92lPKdonLNj370I4yPj6OhoQGf+tSnoNVqceutt2L79u0AgDvuuAMAZG9XkoWq071w\ncgzrOx3wcQlMS1xsVM6OiVAMa1tnFxHTaDSwGfWSFhoB+XaE4ymwFfRqbcwpbVBs0r5Qk7cwerTY\njDg1HhKdeLVmyg2LkWtHJJESlWhy4RdEiR9jPl5XlaIWW+rdDtE79POf//ysbRs2bMDDDz9c9XZK\nbflt3yjuuGIJfn/YXVX9dYGJYAzNNmPR7/h6L/I7KnHxFCwG8QdDIWZGB61Gg3C8fLGxNOELkrGM\nDq1Wfj5BbOKVZeTbEomnwDLS7TDJKDdModQSVc4CLcTc2RFfBCO+KK5a3sjr1xIzbMrZMR6KocVa\n3MlbGZ3kUgC5doRjSViM8iN5AGhgDSUfXoId/IQo/0BoyTygbMbyi6FYiZF8fp58GmaD9NtHaiXK\n+XZdVYNabKl3O1Tp5Bciw74IljdZoNdq0WBhZKVRFoMQgslQPFtzvZBqInk5EXAuUnLlBakGQI6T\nF0uhrCCSTyRlyTW8k6e58pT3HlU6+YWo040Fo2jNODU5TTZK2eGLJGA2aGEq4cj4yo3ye6OGE8mK\nNHkAaLIYS1bAFOwIx1LZN4UWmxF6rQYmkYhbsIWIlDLO1+TT8jV5CZUo59t1VQ1qsaXe7VClk1+I\njAVjWScvZ3VoKcaDsZJRPJCJfsuU/i0FF0/BUqGTX9/pwIEhX9l9hPrxANBqM8Fm0ouurjbotNBp\npLUAFODifAqlVEwyShtQKLVElU5+Iep048EYWmz8RGM57bqQUnZMhGbGK4ZFxmRlYR32SuWazUsb\n8PrgVNGIW7AjHEtmHyKLXWb8219eLGlsKbp8viYvX66RMvE6366ralCLLfVuhyqd/EJkXIlI3saU\n/N7C6BCqKJKvXK5Z0WRBMkVw3hspuU84nspq8hqNBquLpIAWQ2rjEIFIIg1WjlyjpzXlKXODKp38\nQtTpxoLR7ERjo0jFxlxK2VEuswYQ5JoKNPkcJywXjUaDzcsa8Mbg1Kzvspp8vLLsHSmlDQQ7CCGI\nxOXlyUvNrplv11U1qMWWerdDlU5+IcJH8ry8YjXqkUiRqvKyJ0IimrzMyFeAq3AxlMDmpQ14Y2C6\n5PeVPkTk2JNIEWi1M4XHpGCS0cybQqklqnTyC02n4+JJxFPpbC9TjUaDBotBUhplKTtyNf5iyGl+\nnZcnH0/KkjkKuXxJA94d9s2qZ5+nyUtYhVuIFHsEOziZq12BzGIoCdk18+m6qha12FLvdqjSyS80\nBD0+N4tErMmGGBMico01U9ZALlwVcg3Al/Rd1mjBu8PFs2zC8RSsFbwpNLAMnjniljSXEa3AyZtp\nn1fKHKFKJ7/QdLqxIlG31MnXkpp8mZIGgLzsGsGOZDqNRCotmrcuxq2XLMIDfziGNwdnZJuZPPlk\nRQ+RrR9YiRabCR//2ZvYf95bfJ+MHZxMPR6gmrxSLLR7vRIqD6ko84bchVACchZEFRJNpBBPpeEo\nU/NFzmIoASF9stquYB9e14YWmxHbnj6K/3P9KvzJ6pbsd5VOvNpNBtx93QVwsQa8fGYSmxa7Su4r\nlBmWA19qmGrylPceVUbyC02ny10IJSA1ki9mx3hm0rWcM7bKWAyV1bKrnHTN5dLFLtxxxRIczMg2\nWU2+SjloWaMF573Fm9oIdsitQAnQ2jVKsdDu9UpQpZNfaBSbJK1Gk58QWe0KVBbJ83VlauPkAT76\n9kXyu0VVOvEq0OUyY6hMHj7AO3m5k8cmqslT5ghVOvmFptMVk2sq0eQJIeg9PYEHnz2OzUsbyv4d\nn1eeFK33AuRr2ZWudi2Gw6xHIMo7+Zk8efkOOJdFTjPc/iiS6dkOWbAjIrOkAUDrySvFQrvXK4Fq\n8iqAj+SLaPIyV70+f2IMj7xyFttuXIPLlpR38jqtBka9DlxCujwSjlWXPlmI3WRAIJovGfGafOWX\ntVGvQ6OFgccfxSJX8d6wSso1FEqtUWUkvxB0Ok8gir2n+WbaReUaizS5JteOUX8Uf7K6RdTBC1gk\n1pTP1eRrKdc4zAYEMnJNbp58JSmUuXS5zDjvmy3ZVJ8nLy7XiF1Xxz0BJFPVyT5Uk5dHvduhSie/\nENh9chz/8LvDeP74WNFMmIYKInm5Oexy67CHE7WVa/hIfkaTJ4Rk2gtWd4wuF4uhEpOvQOV58rWI\n5B/4wzG8Plh6xS+FUogqnfxC0OnOTIZw00Ud+NbzJ2YthAIAu4kvbRARmRzNtUNu9gsfyYs7+RlN\nvvLiZMXg+8ymkCYEW7ZsQTyVhlajAaOv7rJe7GKLTr7mzy3Is8OQKYGQEInCxa6riVAcR9wBWccu\nhGry8qh3O1Tp5BcCZybCuOmidtx/41pcuaxx1vcajQadThOGi8gOpeBkNvSwyihtAGTeFKrQywvR\naTVgGR2CGV0+t2FINSx2mXFumo/k/3DUgx17Tud9H0mkSjZTKUe1ufLRRAqhWBJHR/0Vj0FZeKjS\nyatdp0ulCQamw1jeZMF1q1twz/Wriu7X5WIx5CstOwD5dsjNfrEw0nLlBTvCNdbkAf6NJRBNoLe3\nN6/1XzXkyjW/PjiSzZvPzZOvZAJZSgvActfVZDgOq1GPo54g0hKymkpBNXl51LsdqnTyamfUH0ED\ny4g6tEVO8ZzvXGTLNUYdQjI0+VrLNQBgN89k2NTqIdLpMGEiFMeZyRD6Rv0IFWTwVFLWAKhel58M\nxbCsgYXdpM++aVDUz8mxIL72+yMV/70qnbzadbr+CT6KF2OxyAQikG9HOJ6ERYbz4uvXiDstpfLk\nAcBhMsAfTWDLli1VL4QS0Ou0aLUZ8ZN9g7hssQuBzNuKYEe0grIGgLRc+XLX1WQ4jkarEd3t9qp0\nearJy2Ou7Tg9EcKeUxPZJAMunsT2545LbgykSievds5MhrCiySq6X5fLXLaLUiGy5RqZfV7DMXkP\nESnkplH6IwnYaqT5d7lY7D45jk9u6podyVeQXQNk0iirjOSbrQzWtTtwdLS6yVdK/eD2R5EiBG9k\nsqr29k/i5f5J/O3j72IyVLyxfS6qdPJq1+nOTIaxUmIkPyxbk5cx8SqxtEGuJl9zucakRyCaRG9v\nLzzBGNrspWvgy2FxgxkrmyzYtNiJQIx/iFRTuwaQJteIafJNFiGS9yOWTOHvdh3ECyfGZP0OqsnL\nY67tGA1EsKbVhtfO8B3Rnj8+hruvuwA3rG3BFx5/F6l0+fkZVTp5tSM1km+xGRGIJkXTKAU4mbVl\nLIy8mvJcjfPkAT5X3p+J5PnyDrVx8tevasEXrl4B1qBDIkXyUh/ltv4TMBl0khZElWIyFEOTlcHq\nVisGpzn801NHcXwsiMM0qlc1bn8Ut17SiX0DU5gOx/HusA/XrmzC565cBn80Mat+UyGqdPJq1uni\nyTRG/VEsaSi+5D4XrUaDDoe5bIZNviYvV66R1jJPqTx5IOPkM5r8WCCKVnv5wmpSuXiRE9esbIJG\no4HNqEcwmpypXZOo7I1ESiRfVpMPxdBkNcKo12FFkwXhWBL/9KerMTAVlvU7qCYvj7m2wx2IYuMi\nJxosDH7w8hm8b3lj9j51mhl4RVa2q9LJq5nB6TA6HGbJC34Wu8wYlqDLC+305Cwk4lMopWvMSqRQ\nOkwzRco8wRjaahTJ52Iz6rOSDVBdnnxV2TXhOJosDADgmx9dh+/+xXqsarFicFqek6fUD8l0GhOh\nGFrtJly1rBFPHXHjhjWt2e9d7OxKrIWo0smrWac7MxnGCgl6vECXi8X5MpG8YIdcqQaQXtYgv558\njbNrMimUvb298ASiNdPkc7GZ9AhFk1XnyUvJrpGiyQP8eWUZPdodZni5BDgZqaxUk5fHXNoxHoyh\ngWVg0Glx9com2E36vMWPTrOhNpF8IpHAF7/4RTz33HMAgL6+Ptx///245557cOTITP6m3O0U+Yz6\nI1jkNEvev0tirjxXQVqgVeJiKICvKxNRZDEUn12TTBMEogk0ZiLdWsJH8rydyVQaaQIYdPK7W21c\n5MQv9w9h1C8940kgnkwjHEvCyRrytuu0Gix2sRiss7x5tz8CTyA61z9j3uP2R9Hh4AOXSxY58evP\nbc5723aZmdpE8i+88AKWL18OjUYDQgh27dqF++67D/feey927doFAEin05K3S6lBXg1q1ummwnFZ\njkysCYZgRyVRNq/JS8uTjyXT0Os00Otq+/JoN+vhjyax5pIr0Gw1QqetrrVgMWwmA0IZTV7IrKmk\nheF1q1vwV5ctwRcef7ekgyt1XU2FY2iwMNAWOe6yRhaDMnT5+aDJf6+3H5947E3sene4qtW7ar7X\nAWA0EEW7fSaoc7L5976TNcDHVenkY7EY+vr6cOmll4IQArfbjfb2djAMA5fLhdbWVrjdbng8Hsnb\nPR6PFLspRZgOx9Egy8mLL4gCMnq5TAlCTjPvUKz2k64AvxgqEE3AU6QFYq2w5+j+XIVSjcDHNy7C\nn6xuwc43z8n6u1w9vpCljRYMTNVPJE8IQd+oH9+6aR1+1zeKX74zNNc/ad7i9kfQ7igtQbpYA7yR\nKuWaZ599FjfeeGP2cygUAsuy2LlzJ3bu3AmWZcFxnOztSqJmnW6akxfJt9iMCMaSJTXbXE1erhMW\n6smLvZn19PRU7RxLYTPpEYwlsefNgzVLnyzEauSP0dPTg0i8sknXXK5Y2oCzk8Uj71LX1WQojqYS\nLRmXN1pkRfJzrcl7AlGkCfC+ZY348Lp2eIKVyzZqvtcBvsdDRzknb2bgrSaS5zgOJ06cwMUXX5zd\nZrVawXEcbr/9dtx2222IRqPo6uqSvb0cuf+pvb29sj/39/dX9ffz+fPQhB9njhyUvP/Le/fCrk9j\nPBgr+v3BgwfR29ublWvk/B69TguthuCFl8rv39/fX9H4Uj6/+vLLMGgIPBGCNrtJkf//ydGhbKXL\n3n1vgcQ5WX9f+Nl98hDOZTJiSp2Pwr+fCMXQbDEWHW/i7FGczUTyUo4/1/fHr3a/gfUdDmg0GgwN\n9OPMueGJSwPQAAAgAElEQVSKx+vv71f89x48KP1+q/Xn4+c9GB84WfL7wVPHMDA6jrKQMuzfv598\n4xvfIDt27CD/9E//RP7xH/+RDA4OknvvvZd4vV4yNTVFHnroIUIIIalUStb2Uuzevbvs9wuda3t6\niT8Sl/U3n/vvd8iBIW/ZfX7XN0IeeOao7N/zwR+8Qv543EMODHlJOp0uud8756fJX//3O7LHl8JN\nP3qNfOFXB8iT7w4rMv6v3x0m2587Rggh5PnjHvIPv+2rarx0Ok2u7eklPk76eXzk5X7yH6+dLfpd\nPJki73t4D4knU1X9rveKb79wkvzirXOEEEL+eNxD/u/vDs/xL5q//Nm/v0aGvFzJ70+OBcgnH3uD\nEFLad5adadu4cSM2btyYfYLEYjEsWbIEt956K7Zv3w4AuOOOOwAAWq1W1naKfKKJFOKptOz6LC6W\ngVekmJHcrlACH7qwFU8dduPsZBhf2bISH1zL5/B+r7cf69rt+JPVLdnxa50+KeAwGXBqIoRPXVb+\nDbFShBRKABgLxqqWhTQaDZZmJJYNi5yS/mYyHEd3u73odwadFm12E857OUkroeeavlE/bshcJyzt\nfVuSZDqNyXAMbWXmmlwsA2+t8uS3bNmCG264AQCwYcMGPPzww3j44Yexfv367D5ytyvFXOt0wWii\naO7qWDCK3x8elXyMQjumuTgaWEZ2ZofLbCh5IQh2VNLtCAC2fuAC/NutF+NL1yzHc8f5GirRRAq/\nOTSC/okQgIwmr8BqVwG7SQ9/JKGYJm8z8vVxenp6MqUTqp/gXdpQPO2x1HU1EYqhuYQmDwDLG1kM\nSpx8nUtNnosnMTgdxppWGwCAZXSycvwLmet7vVYUs2MsEEOjhSmbkeY084uhSJl5MVUuhpprnjw4\ngh++ciZvWzSRwt//5jB+sm+w4nHlZtYIuAqaet//zFHsH/Lm7ROu0glfe0EzDgx54YsksLd/ApFE\nKq/HbK26NhXDbuZzx5VYCAVkUigzefJ80/TqnfyyRotoOQJCCP7mlwfwnd2ncG6aKznxCgDLmyw4\nnXmo1prHDwzh6SPumox11B3AqmZbNtebZfTgZHQXqyeqqTgK8OUMctMni2HQaWHSaxEss15FlU5+\nrnNnJ0NxHPcEs58JIfjmH/lerGLpTrkU2iE3R16ggc1fFXdmMoyxAD8RK9hRaT0WAQvDr8R76dQ4\nnj7iwfWrWzAZ5o+xdetWTIZjaLQoleJogIXRwVrD1oL54/MplFu3bq2JXAOUjuRzrytfJIFT40E0\nWAxot5vKLoJb1+7AEYltAeXcH5F4Cj/eN4hDI/JbDha7R/pG/djQ6ch+Nht0VTn5ub7XS+EJRPEX\nP3ld8v7F7Bjycuh0il9rLrb8gihVOvm5ZpqL48xkGLEkf/HuG5jG6YkQtn90HdLpyp/wU1wcjWwF\nkXxBmtVkKDaremQtNPMb17biiQPDOOoO4JaLOzEVmnmwjIdiaCkTiVaDw2xQLIoHeLlGiJRqVQRt\nqYS0x/FgDB0OMz535TL8+yc3ln2Ire904Ig7IFp2Vi6/PzIKrYaXGmvBkdEAujtm5hYsjA5cHWvy\nB4a8+MTP3sQnfvYm/vXFU9nth0f9mAjFRRu3l4MvKS4+x+IyG8qmUarSyc+1TjcVjiOdJuif4G/i\ntwanccOaVpgMOn6FmshEicAsTb7CSN6VE8knU2lMc4nsIibBjkoWQxVy5bJGjAdj+MCqZnQ6zFm5\npqenBxPBeE1kjmI4THoYEsoV6bKa+JLKD+/4HnyR2pRO6HSaMBmOz3rg515XHhn6v9NsQLPViDOT\n4pKN1PsjmUrjv98ewt++fznGguLNKQopdo+cHA9m9XgAMDPVRfJzfa+/fc6LjYucuO+GNXj6iBvJ\nNO/Uj2TKP1d6rwNA/0RIUp0qJ8vAV6Z+jSqd/Fzj5eJY127HcQ9/ot8Z8mLTYhcA4akrXbLJZapS\nTT5nBn4y43hnR/LVT4wyei2+vGUlPnXpYjRm5gGEJevjwahikfy6dge6XbUvZyCg12ph0usQgBEN\nFgZ6bfW3jV6rRYfDnG0SXgy5+v/6TkdFskopXjg5jnaHCdevasFYDerMTIfjiCXTaM956zIbdIgl\nU1WVNhDjvJfDc8eUWWV/ZjKMS7qcuKjDgXa7CSfH+IfsYbcfGqDie50Qgv7JMFY2i0fyzjKJFYBK\nnfxc63TTXBzvW96IE2NB+CMJjPgiWNfGRy9SUp4ECu0Qsmvk4mJnak5PZNqFCU6+mto1xfjY+g4s\nb7LAoNPCYtTDx/Fatlh2SDVs6HTgH2/9gCJjC9hMemy+4WM1LZ2wrHG2Lp97Xcl18hs6HeiT4OSl\n3h+vnZ3CR9a1wW7SI0WIrAYxwOx75OR4EKtarHnZYVqNBka9eHXOUkixpffUBLY9cww/e2OwomOU\nu9dzq8JuWuzC/vNeJFJpnJ4IYW2bTXQ1qkCx+TcQIumt0cWWDxxV6eTnkkQqjXA8hc1LG3B8LIgD\nQz5c1OHIpkE5zYayr1blqFSucZj1CESSSKXJLCcvoERrviYLg6mMJMElUrMqKNYTNqMe/ROhmqZp\nLm0oX1hsLBhFi4zjre9woE/i5KsURv0RdLlYaDQatNpMVevyvJO3zdpe7eSrGEO+CD575VI8e8yD\nx14frNm40UQKY8EoFrv4Bj4bu5zYP+TDqfEQFjnM6HCY4ZORaJHLmckwVjRbJaVLO0UqUarSyc+l\nTuflEnCZDbig2Ypz0xxeH5jCpRmpBuCrxkl9uhfaUalco9dqYc1kiEyEYmizG7PVI2fy5OXXkxej\n0cJgMhzDd37wIzRbjUUrKNYKpXOZbSYDdr99pKbzCqtabHjnvC8vxznXjnGZRdeWNLAIx1PZB3kp\npN4fuXVTWm1G2bp84Tk5ORbC6iJOvprJVym2DPs4bFzkxCMfvwSPZxID5FDq2jo3zWGR0wxDJoDb\nuMiJQyM+HBrxobvDkYmwK7vXz0yGJE26ApnGIQtt4nUumebicLEMTAYdFrtYPHvck+fkXWbpck2x\nsSvJrgH4NMppLo6JUBxLGyyKZNcU0mQxYjIcB0cMikk17xV2kx4+YqppJH/tyiZMc3HsG5gu+r1c\nuUaj0eCiDrskyUaMaCKFYDSZPW+tNlPVuvyp8SBWt852XNVOvoox5OV7MDRZjfjqB1bim88dryrr\nReDMZBjLcyZGnSyDDocZvz44iu52e1X3ev+E9OZAYpUoVenk51KTz81lX9Nqg06jybuw+frP0l7h\ncu2IJlJIpEjFueBOMwMfl8BkKIZljZY8TZ4QUvGK13I0WRlMhWO49sY/UyyzRkDpmt9Wox4cmJpq\n8nqdFn937Qp8f29/NitDsIMQgvFQDK1WeQ+V9R0OHBaRbKTcH54An9kjvH1VEsnn9w9OYjwUK9qb\n2FLFqlcxW2LJFLxcIpv2esPaVrTZTfjPt6SXei51bZ2ZnJ39srHLifNeDt0d9orv9ezYEiZdgZl7\nuxSqdPJzSe7k6Lp2Oy7pcuZlY1T6dOelGkNFzSqA3Eg+hmWNbF4kH0+lodUi+9pZKxotDCZDcUUn\nXd8r7CZ+PqHWNeuvWdEEh8mAp4/kZ3/4IwkY9VrZ3bo6HeaK0h0LGSkocdtqr06TPz0ewooma9HM\nJLNBr1j9mlF/FG12Y/a4Go0GX96yEv97SHp5kVKcnQzPqhW0qcsFC6PD0gZLJuFB/r2eJgQDU5zk\nSF4obVAKVTr5SjT5/ee9+OZzxyXvX1KTz9HNb7qoHQ9++MK878X0s1xy7ag0s0bAmbngJgoi+d7e\nXsWKhzVZjJgKx/HHV95ULH1SQHFNPvMG1VrjRVcajQZ3bl6CPxzlywYIdowFK1s81mjh357KIeX+\n4JtVzKywbbUZs6ukpZJ7Tk6OB7GqRGTKMtI6jBVDzBZeqsl/e1jSwCIcS8IvMdgqdW0VyjUAcOWy\nBtx/41rotBq4zAbJE6+5doz6InCY9ZLf2hekXFMJr5yZrEmfzCkujoZMFolBp81GgAL8Yij5M+68\nDFS5oxRKG0yE4ljSwCKSmMlNDivQexWYcTgc0aNZYblGaWwmPTQgVT1oS9FmM2G64ME/HorJyqwR\naLTyb0/VUtisgs+uqfwN4eRYCKtbZ0+6AryTjyikyQ/7OHS58stBaDUarGiyZgvoVUI4nsQ0F0en\nI39sk0GH6zKVV+UkWeTSX+QNoRxmgw7llhmo0slXoskfGPbJygMupdNNh8tH3FI6uQjk2jGdkWsq\nxWlm4A5EkUil4TQbYNLzk11btmzhF0Ip0LVJcDjOtsVoVTiSV1qTtxn1aLWbFOkh68pZBS3YMRaI\nVjSPIUx2l0PK/eEORPMcWKvdiLFgVFZ/5txzct7LYVnjbD0e4J1UpZG8mC3CpGshF7RYJRd0K3Zt\nDUyGsazRUvZ6kCPN5tox5OWwxFX8/6oYGo0GDnNp36BKJy+FPxz14IUTfGncUCyJU2Mh2Ys9ijHN\nlU9ztJn04BIpJGXO7ldat0agwcLg1HgITRa+VLHVqM/aq6RcMxmOYywUq/tI3m42KFbK2J5pFJ5b\nd2Y8VFnPWgujAwGR3Hu3FCMFvUUtjB4GnRaBaGXjhmPJkvKDhdEhkqj+3ivGsC8yK5IHgAuapTv5\nYpyeCGF5Y3nN3GHmz6swqS6VyXDpVo+l+OWdl5f8TpVOXkyne+3sJP71xVN49JWzSBOCQyN+rGi2\nyIomyuXJl4vktRoNHCa9pJoWuXZMBKubvHSxBpydCmfHsGScfG9vryILoQD+5k0TgolgVPGJV6U1\n+cuXuNA19pYiY+u0muw6BsGOSksaazQaNLJMXpnnQqRo8rxck+8c+Qwb6ZOvueeES5QOJKpJoRSz\nZdgXQWexSF6Gky92be3tn8TlS1yzd85BOK/+iPgDLNeOSqrNFsrCuajSyZfj1HgQDz57HD23bIBR\nr8X+814cGPLi/SuawMWTVdfQkLJgyVnBrLsnGK1q0s9lZpBKk2yEYDXqEM5G8rVfCAXwDqfJwsCI\nZM0zd95rjHodHJrqs1ZK4SyoJMg3J6nsfDdZjZgUWRBVjnA8iWgilZ1bEijU5Z875sHnf3VAkoQT\nKdPInTUoU1M+mUpjLBhFR5Ga7Cua+Xr+cqNsgC8N0jfix3WrWkT3LSzzLYVKS4qXor7vvBKU0+me\nODCMT1+6GOs7Hbh5Qyd+0zeKA0M+XLbYBaNeeiuyYjpdKk3gj/IrXsvBd2oSP/G5dozJXP1YiHDD\nChkbFkaPUDyZ0eSVa83XZDViSUv5iKcWKK3JA8quvxAyMQQ7qmlO0mRhyuryYna4M5Ouhem6fIYN\nH8mfHAvi4ZdOYywQxeuDxRdz5Z6TcuswWKbyFoDlbHEH+DdIoUFJLhZGjyaLEUPeiOgxCq+tZ496\ncN2qZknprWIlBwRy7ai1k1fmzp7HjPqjuD4z+/2hC1vx6KtnkUoTdLfbYTXqEIolK+p1CgCBaAJW\no75suy4gU+RfZiQ/FohWVTPdbjZAq0FOJK9HOMbfWErJNQCfYROvwepCtcPXNOKvCUIIn0JZqZO3\nGqvKsCkm1QD85Ovrg9NgGR1+9NoA/v76VUimCX7x1jm8b1ljyfGS6TQSqTSMRZwtwPd5rXTitRzD\nvuKTrgKCZLNMRFvPhRCCp464cd8NayTtL1Y8rBjT4RiaaCRfnnI6nTsQzU4o2UwGbLmgCatbrDAZ\ndHmOT4xiOt1UOD7rFbcYTomRvGBHKJZEmkB2A+9ctBpNpuY4f/FYczR5JfuvNlmMGBs4Jb5jlcxV\nH85a4cx09+nt7YU/ystblQYbYrnyYnaM+iN55YAF3r+8CXaTHq8PTOP2S7vwwbWt+OCaFgx5I0Xr\nwQjnJBJPwWzQlVzIx6dQVjbxWs6WUpk1Ahc0W3F6XFyXz722jrgDSBOC9TndrcohNZtOsCOe5Asc\n2kXUADksqEg+TQjGglG05Widn79qebagk4XRV5VhI3XBkpwFUcDMEvNKV7vOHJfJToAKTr4BfGOD\n3JzoWnL5Uhc8fUHxHRc4woO/AZm3tiqkuUYLg3NVrPlwl4jkV7fa8MCH8hf36XVa3H7pYvzX2+fx\nLzd1Fx2PE2ktaVaoz+vgdBhLGkpH6Re0WPG7PnkrX3tPT+BP17RKvhflrosRfEgti/mpMpIvpdNN\nhuKwGQ0w5UwAtdlNuKiDfypbjbxOLYViGrDURttOifmzgh3V6vECD374wmx/TQujy2rypV7Pa8G1\nK5vx8N2fVWTsXOpek8/kym/ZsgWeYHXSnNjEq5gdw/4IOmU89G9c24o3i+jy2f7BInM+rKHyKpTl\nbMmt9V6Mlc1WnJKQYZN7bU2F47ICIqmVKAU7aq3HAyp18qXIlWqKYWFmMk4qYVokfVLAVaZw0XFP\nYNaM/1iVN73A6lZbdr4gN0/eXeL1nPLekZtdMxaIVZVJJTbxKsap8SAuaJG+4tLFGpBMEwSjxZ1Z\nOF46swbg5RolIvmzIk6+w2FCIJqQ9fbujybKLjwqxCVx4lWAOnmJlNLpxJxZruMTo5gGPBaMSnLy\n5dp1/f1vD+OYm5c3BDv4SL62Tliwdc+ePXAHoorJNcDc9+GsFUpr8v6MJu8J8EW1KqUxUzOoFOXs\n8HFxBKPJornlpdBoNOhwmDDqz8+hz2ryInINq0A9+ekw33qynMPUajRY4mJFpa3ca8vPJeA0S3fC\nTokTr4IdU+EYdfLV4A5Eyzp5C6OveJY/Ek/huWNjuGZlk+i+LpaBJxDFD14+g48/9gZiSf6YgWgC\nY8HYrEnZsUA0Wyq1VliNvK1cir/YbWUWU1CUx5nT+9dTMG8kFxdrQCiWrKhm+omxIFa32mRrwh0O\n8ywnLxCOJ8umGypRu0YoAyymnS9rtOBsme5chfCRvPSpTLlVZ2kkL5Fcne7pI+6sPun2l5dr5ETy\nhRrwkweHsbHLKanxbrPVCH8kgXAsiVSabxAA8BoigOyk7IwmX/nCmFIIti7v3lT2/6QWzHXP3Vqh\nfJ58RpOvMl1Wq9HAxRpKRvPl7DgxHsSaIt2bxOAj+fyc8zxNvoxcw+i0SKWJ7FIfQGlbzk6FsbxR\n/F5c1mgp24IRyL+2fBF5kbwgzU6GYnj++Fhe6Ypc8jX52gZ0qnTyuTxxYBgvnZoAICGSN+oqqvkR\niafwX28P4XNXLpW0v8NsQO9XrsU//ulqXLLIgeMePv1MqIpXqOHxGm1tT7yF4dcEuP1RdFA9fs4R\nFs0QQjAWiFU9B9MkItmU4uRYCGvaKnPy7hKdo8qVNAB4ucdchWRTjLNFygAXY2kji4EpaZlIyXQa\n4VhKViqz02yAP5LExx97E///CydxRKT1II3kJZKr0/mjCewf8gIQn3itVJP/30MjuGSRQ1IUX8ia\nVhtOjPEaPN8o2ph9be/p6cl2CKrm9b0Ygq0vHziqWGaNANXkxTEZ+FvxuRd7Mc3F0WSt7kZvtPL9\ndYtRzo4TY0GsKVESuBwdDjNGCiL5mf7B5SN5oPLJ11K2iGXWCCxv5MsblEOwIxhNwmrSy6pEqtdp\n8V93XIZnv3gVPtLdhgMZX1TIjCZPnbxsgtEkDgz5kCYEHpFI3spIXwyVyztDXtx4YVtFv29tmx3H\nPIKTD+PSxa68SN7LJWDSa/PSPmsBv/ArCW+cKC7XUMTRZBarjXJ8nnuxDkpyaLQwmJK56jUYTWAq\nHMdiGWVuBYpNvApE4inREgCsoXYZNoQQyZH8IqcZ48EYohLeInyRBBwVzF2tbLbCqNdh4yIXDgz5\nyu6rxMSr6HvHE088gePHj8Nut+Mzn/kMnE4n+vr68OSTT4LjONx5553o7uYXQcjdrhSCvpVK871L\nW+1GvHPeC5NBV/a10VKhJl/N6/XKZgvOezlEEymcmQzh5g0d2H1yPGvHcU+g5t2IAKF2TQoaayPV\n5CWitB1OswGm9i60hSaqHosv81w8ki+044kDw7huVTMGpzlc0GytqGZ+h8MMt5+vNy9MdgrnJBxP\nijoutsI+r7m2/PT1AbTYTNi8tAF6nQYuCZluep0WnU4Tzns5rCoxFyHY4Y8k4KxiJeolixx48Nlj\nSKbTsx7iW7duBSGELyn+Xjv5j3/84wCAgwcP4rnnnsMnPvEJ7Nq1C9u2bUM4HEZPTw+6u7uRTqcl\nb1+3bl3ZWe9P/uxNfPqyxfhod3tVxoVi/Kz+ZUtceOqwWzQXXM5iqFzGQ5W1agP46oZLG1i8fGYS\nLKPDkgY2L+VKifRJAGD0Wug0GgxMhWd1t6HMDS6WwYmxYE3WRDRZGMmldP/jtbP4/eFRXLOyqSKp\nBuDvHYNOA18kMcu5RhJ8WYNyVJNGKXBw2I/9Q4P4zOalkvujAsLka2knL+CLyMuRL8TJMmizm3By\nLIR17fZZ33OJFDTQ1LxYoKR3wmQyiaNHj6KlpQVutxvt7e1gGAYulwutra1wu93weDySt3s8nrLH\nc5oNGPKJV4crhaBvBaMJ2E16XNrlwp7TE6JOnl8MJa92TTSRQiSegktCzZpSrG2z46nDbqxstmaz\nLAQ7+MwaZWqxW426TKNjZSN5qslLw2E24MDAWE3OdwPLYLrExGuuHcl0GqF4Cmtabfjp64NY3Sp/\nXkmgMI0yT5MXkWvMBn1FaZS5tni5OO68Ygl+/NqApMwagaUiurxgR7WRPABs7HIWlWx6enoU0eMB\nibVrvva1r4FhGNxyyy04f/48WJbFzp07AQAsy4LjOKRSKVnby/HBNa04PlZ9vRN/NAm7yYBNi12I\nJdPSInmZK17HQzE0W5mq6sqsbbXhd32j+PRli/lCVTnLoD0KZNYIWI16RGMJyQ2DKcriYg2YiKIm\nD12bSY+ghGvZH0nCbtLj/35wNcyMDpcvaaj4mB0OE0b8kVlRqlh2DZDfzLuYnCEFXySBj3a3o4Fl\nsFRGZcllDSx6T4tLZNVG8gCwscuFZ4668VeXL571nVJOXtL/5He+8x3cfPPNePTRR2G1WsFxHG6/\n/XbcdtttiEaj6Orqkr29HKMDpzCdkSx6e3vzojQpny+++GIAfCSfigRx9J3XsdjFot1hKvv3FkaP\nYDQu6XiCTvf8K2/CmI6K7l/uMzdyCgT8BM3b+15BLJlCLJnC1q1bcbj/PLzDA1WNX9Jeox4NRtRs\nvFKfhfOh1Pi550Op8QFeN1VyfKfZAAJg4tzpqsezmQwIZqqMljsfL+x9DYZ0AnqtFv/nulU4ceCN\nin9/u92E1949lv28ZcsW9Pb2YnRsMivXlPp71sDXlP/2rj248yfSj3/xxRdnP/siCRx55w00+ftx\n6WKX5N8/NXgcA5lVr8W+F/BHEph2D1V1viPnj2L/4FQ2X174fuvWrZgKx5Hm/FWNXxQikXPnzpHt\n27eTVCpF7r33XuL1esnU1BR56KGHCCFE9vZS7N69mxwa9pE7f/G21J9Wkj8e85D/+7vDhBBCnjo8\nSvongmX3T6fT5IrvvEQSyZTkYzx9eJTc99SRqn5nLJEim//1JXJyLEAIIeRDj7xC3P4IIYSQv/rP\nt8jhEV9V45fiC786QP7ht32KjE2Rz6/fHSaXfvtFcmqs/HUqhWEvR2760Wui+71zbpr8f/+zv+rj\nEULI4/uHyLf+eHzW9jt+8bboNbzjpVPksdcHyJ/9+2vkqu/uIVwsKevYkXiSXPXdPSSdTsv6u9y/\njSbKH/PBPxwjvzk0Inv8Qm796Rvk4JB31vZf7T9PHnrhRMXj7t69u+h20Uj+Bz/4AbZv347f/e53\n+Ou//mtotVrceuut2L59O/75n/8ZN954IwDI3l6OBguTjeQrQdDpAhlNHgA+2t2OFU3ldTqNRgOL\nUYeQBG1QeHqOBasrJgXwk6A9t2zI5tk7zXwdk56eHoyW6FFZCyxGPZL+6jM5xKCavDQEvbeaujUC\nNpMewRJNt/N07EiiqvmkXDqdphKafPmyBgAv1/y2bxRdLjPWttpwcKR8qqGAYIuXi8NhNlQkm5oM\nOqxqseLgsL/o94Id/mj1mjwA/OXFnfjPt8/nbevp6cFkaI40+bvuumvWtg0bNuDhhx+uensphEbE\nJCcdqxIC0aTsmiyCLi/1ZI4HY1jRLF3/K8UVS2e0UKGOSZxoEU+RmlxYxWhgDdDEa1e3mlIdTtYA\noxY1mSPh6zDxPYvL1aHxcnG4ZCzTL0ep+jVSFkOZDXqM+qN44ENrsf+8D2+d8+LKMt2mCvFVOSl6\n5dJGvD4wlXcfzjoGV70mDwA3XdSOx94YRP9EKG8B5ag/gvevEK99JZd5uRjKzOig1WgqLhYm5M4G\nogk4TPJuGAujl1RuWNCAx4LRitu0lcLF8lUq//z2z6CzSK/NWvGVLStxzy3XKjJ2LjRPXhqLnCyu\nWtlck/Ot02pgNhQvnZ1rh49LwFmjSL7NZoInEM029hbOiZQewg6zHhu7nNjY5cLlS1x461zxvrGF\nCLb4IuK9lcuxeVkD3sipib9/yJutpSPYUe2DRMBk0OH2TV342RuD2W1bt27FsC+iSDrzvHTyAB9l\nViPZAJVG8jpZufJjwdqXHHBlWsGNKCjVAADLiPejpbx3tNiMeOjPL6rZePbM5Gs5vJF4zeQaQZKJ\nJWcKjRFCRDtDAcCHLmzDd/9iPQBgXbsdI76IrN6ovkgCTgmLn0pxYZsdk6EYxoMxDPsi+NITB2fV\nmamVXAMAt1zSibfOeTHknck0HBHpSVsp8/YOb7CUzvMVozBPXg5SV70KOt24ApG8w8xXrvvV088r\nvlBJDVo2QO0ohtVYXJfPzy2vneMCALtJj0DmmL29vUikCDQADCLBRG5PW71Oi0u6nHjnfPE6L7kI\ntlQbZeu0Gly2pAFvDE7hp68PAISXYgU7UmmCUDQJm0x/UgoLo8dVyxuzNn57x/eRSJGaPXBzmbdO\nvrHMYg6pCHnycrDIqF8TTaQQTaZrrpkLC6JCYBSN5CnqRkquvC8Sl1U6Vwy72YBATocoLp6EpYIm\n8bT7Y0AAABbQSURBVJcvacBb58SdvIAvM/FaDZuXNuA3h0bx6pkpfKS7DWPBmfmFYCwJi1FXdU2h\nXFY2W7PlxUMwotNpVkSanbdOvsHCYKpCuUbQ6SqJ5KUuiNqyZQvGgjE0W6tvsF2Ik+U7vDcuXoVO\np7KrUdWgZQPUjmLYSkTyhZp8LaNHu1GfdfJbtmwBlxAvTlaMjYucODgsnmFTK00e4HX5I+4AbtvU\nhRVNFoxlIvktW7bAF4lXVJysHCubLDgzyZeeuObDNysW0M1fJ1+DSD4QqySSL15TnosncdcT7+Y9\nAMYVKjnAR/JxjPqVmYihLAysRj2CsfJdibxFas1UAx/Jz9wjUiZdi7G8yQJ3IJrtmiZGtZo8ALTa\nTNi6ZSU+sWkRWmymrFwD8CuDa/3GviInkldKjwfms5OvIle+WJ68VEpF8v/99hDePOfFee/Myrix\nYKzmejzAp9JNhRMY9oYVrxBJtWzp1JsddpOh6LUs2EEIqVnGiIAtJ5Lv7e2VlD5ZDINOi06nGYMi\nDT1qpckLfOqyxbAwerRYjVm5pre3l4/ka+zkmywMUmmC6XAcz73yFjoVutfnrZOvVpOPJ9NIpoho\n9btCimnyk6EYHj8whLWttrw84HGFKkQ6zQyGfRyMSMKor20decrCgXe4paXHYCwJk14rOikqB4cp\nP5KXUoGyFCuaLNluaWLU+mHVajcWRPK1yZHPRaPR8DZOhhAEszAjeUGTf7l/Ev9dsEKsHFu3bkUg\nmoDNpJetlwvlhuPJNJ495sGbg9P4wctn8NHudlzS5cRopjomr8nXPrMG4HOG0wRYs6i55mMXQrVs\n6dSbHVZT8bdSwQ4vV1upBhBW2s5o8uEKJ14BYGXTjJwhkEoTfOrnb2XtymryXLymTr7RwqcxJ1Np\nbNmyRREnDwArMjZqbc0Lz8k35qRQPn9iTNZMO8BHKXL1eIDv8xqIJrDtmaN4/MAwfvbGIM5Nc/jM\n5qXoLFjRp1Std71WC4dJr3hbPoq6yZ14DcWS+Pmb5/K+93HxqicrC7GbDPDnRvISukKVYkUzH+Xm\ncswTwKnxUF6UnSYE/mhtNXO9VosGlsFEiD9Ord8UBFY0WXByLIjJcPV9fUsxb508r8nzjY3fOe+F\nOyC9vnxPTw8CkQTsFSwPtxr1eH1gGuFYEv/xyY34909uxM8+fSkcZkO2lCrA63SDU2EsaZDfKk0K\nDjOD88cPKjJ2LlTLlk692ZGbQtk/EcL/vMO/DWfrvdRgsrIQe04k39vbi3A8BdZQWW75yiYrzkzk\nR/JvDPCrUqdz+iCHYkmYDbqaL+xrtfGSDa/JKxTJN1uxb2AKxlRUsYWJ89bJswYd0oTgmCcIQoBR\n/8xyaSkEYknYKzgpS1wsblzbim9/7CIw+vz/Hr42B+/kYym+VZdSr1gu1gCrprrsIsrCxmbUI5TJ\nrnEHovBFEtkSt0Cmbk2NF98UZtdEJKx2LUW7w4RwPJmXd//64BRcrCFvvk6pKLvFZsymUZ6eCGGZ\njBr1UlnRZIGXSyh6r89bJ6/RaNBoYfDH42N4/4pGmAw6eLny6WACW7duRSCSgK2CSL7JasT2j64r\nmvbV4TDBE4ghTQg6127E0gZLRf0wpfBXly3G/X9zmyJj50K1bOnUmx0244zD9QSiSBM+4yw3t7zW\nzrFWefIAoNVosLxxZvI1EE3g7GQY165szs7Xbd26lZ9bUMDJt9pMGAvGcMX7rsbAVBgXtlXWGrEc\ndpMBLVYj3rd+dc3HFpi3Th7gc+VfODGGyxa70GE3wR0o3g2+GIFYsuavVyaDDlajHhOhGM5MhmX1\nkZTLtRc0o6nCvrEUCpAv13gy905uWrKPS9R0tStQLE8+WVEKpcCKZiv6M5LN2+e82NDpRJvdWBDJ\nx+FQoBxAi82I8VAURz0BXNBsVSzTbXmTRdH1MPPbyVsYTIbjuHSxC+0OU1YqEUPQ5CuJ5MXodPA1\ns3sPnhCtT18t9aYBl4LaIZ1aa/KhjMN1B6LQgM+omdHkFZBrTIa8PPmIhP6u5VjZbMHZTIbN64NT\n2LysgV8omaPJ16L3ajFabEaMBWL47asHsaHTWfPxBe7cvATn9j2j2Pjz28mzDJY1WtBkNaLdboK7\nSK3qUgQr1OTF6HCaMeqLwBNBTerIUyhKwRp0iCXTSKbT8ASiWNLA5kXy3hqXNAD4eYBwLIV0Zv5M\nSgXKcqxssuLQiA/PHvNg39kpXLm0YVbxQiXeSABerhkPxTAYItjQ6aj5+AKbulywLURNHuBntzdn\nivh3OMwYlSjXbN26Ff4Ks2vE6MhE8tNJA1YqHMnXmwZcCmqHdGpph0aj4UtnR5PwBGK4sM0OLxcv\n0ORr6xx1Wg3MjA6hWJLX5CssayCwutWGRU4zXjs7hb/Y0IklDSyaLEZMhWc0+VrUrSlGi80Itz+K\n0age6zuUc/KAstdW7b1gDfmryxdDSKhpd5iwb2BK8t8qFcl3Osx46dQEEimCZmvtowcKpZZYjXoM\n+yLQazVY5DJjOid5QYnsGmCm3LDdZKi4rIGA1ajHd25en7etgTXkFS/0RuJY2lj7VOYmKwNfhM+g\na1CgLd97xbyO5I16HUyZC6TdbiraWqwY1eTJi9HhMOGd8140GZKKdWwSqDcNuBTUDunU2g6byYBT\nEyG02U1oYBl4w3H09PRk69bUqvVfLoIuz9euEe/vKpcGCwMvx7cH7enpUSyHXa/VosliRItO+hqd\nSlHy2prXTj6XdocJ7kBEcq68P6rMie9wmBFPpdGmzBooCqWm2Ix69GecvIs1ZLstTYXjMBt0NXfA\nwjGFlbbBWDLbDKRWGPU6MDrtTOaQP6rIynOAl2yWWuu7D3LdOHkLo4dRr4MvIp4r/3df/go8AWWW\nCbfajdBpNLhm/aqaj11IvWnApaB2SKfWdthMepweD6E9E8lPc3ye/Ig/qthCPofZAH80gSvedzWm\nwnFFqisKZU/u+vKXMeSLYKlCK8+/smUlvnTT1YqMnYuS11bdOHlAumQzEYrBbtJnpZ5aotdq0Wo3\n5nVZp1DmKzajHqcnQmizG+FimWwkP+LjFGtSIUTyZyZDWNLAKrJcv4HlCxgO+yJosjCK3OsAcPEi\nJ6wKyL7vJXXn5KUsiPreT36hWJQCAN+7ZQOmT7+r2PgC9agBF4PaIZ1a22E16hGOpzKavAHTHK/J\nj/iiitUvF1oAPv3qAaxqUSYYarQwmArH8f2dj2O5wllu9X5t1ZeTd5jglrAgKqhwb9SljRZoFZ50\npVBqgVCJtd1ugtWoRyyZRopoMOKPKHaP2DN17Ec5gguaa18KAEA2V95PTFimQGaNmqgrJy/kqItx\n4WXvV7wBNtWApUPtkE6t7RCkhja7CRqNBg0sg0997vMY8Sno5DORfNjgwGqFInlh1WvbqvWKlhcB\n6v/aqisnv8TF4uR4UHS/YQX7JVIo9YTNpIdBp8nmeTszGTbDPuX6B9tNevgjCfRPhHCBUk4+E8kP\nTHGKVIdUE3Xl5DctdmHYF8Gwt3zfxwMnBxV38vWu0wlQO6RTj3bYjHq02UxZebGBZfDj/3kSgWgS\nzQoVwLObDDjuCYLRpCpq3COFRpbBRCiOsxMBxZ18vV9bdeXkDTotPrimFc8c9ZTdL0gYRau6USj1\nwvImC268sDX72cUaMEVYtDtMipXJtpv0GA/F0KGgVN5gYXDE7YcZCcUya9RCXTl5APiz7nb84Zgn\nWwCpkGA0Ab3BqMhy7VzqXacToHZIpx7t6HCY8TdXLc9+drEMjB2rFA2ChOh989qlih2j0cIgEE3i\n4hUdih1DoN6vLdEE0MceewxDQ0OwWCz43Oc+B5fLhb6+Pjz55JPgOA533nknuru7AUD29kpY1WIF\na9Dh3SEfNi12zfp+xB9Fp9OseMkBCqUeaWAZHPMEcNNF7Yodw27i3cpqBdeSNGTaFi6nerwoopH8\nZz/7WTzwwAO45ppr8MILL4AQgl27duG+++7Dvffei127dgEA0um05O1y2vgVotFo8JHudjx91F30\n+2FfBNHJkYrHl0q963QC1A7pqMEOF2tALJlWNJI3G3TQazWYHjiq2DFMBh0sjA6n97+q2DEE6v3a\nkizXWK1WJJNJuN1utLe3g2EYuFwutLa2wu12w+PxSN7u8ZTX1MW4cW0r9p6eBBdPzvpu2MfR3qgU\nSgmECFjJxASNRoOeWzagUeHGZi02E1wa6T0mFioaIjGs/vGPf4wPf/jDCIfD2LdvX953V199NVKp\nlKztK1asKHqcF198Eddff73o7/nqrw/hT1a34CPd+a+d//zHE1jTasMtF3dKMYtCWVAcdQdw53+9\ng1/eeXndl+YIRhOwKZS9U4+U8p2SIvl33nkHHR0d6OzshNVqBcdxuP3223HbbbchGo2iq6tL9vZq\n4SWb2W8EfP6vMsu1KZR6pyGTkNChgnuEOnhpiDr5M2fO4NixY/jIRz4CAGhra8Po6Cg4jkM4HEYg\nEADDMLK3lyNXA+vt7S36+eoVjTg9EcJv/rgn+73bH8Eptxc/+d63Rf++2s/CNqXGB3idTsnxc4+h\n1Pi9vb246667FB2fng/pn4/tfwMd/pPZbk1K2lN4bpQ43l133aXo+L29vXl6uVL2CMeodryiEBG+\n9KUvkfvvv598/etfJ4899hghhJCDBw+Su+++m9x9993k0KFD2X3lbi/G7t27xX5Sln95/gR59JUz\nZPeJMXLXE++S67+/l3z3pVPku9/dIXmMStmzZ4/ix9ixg9ohFWqHdN4LOwhRjy31Ykcp3ylZk3+v\nkKrJAzP64qYuJ/58fQc+cEEzXRhBoVAWJKV8Z10XSl7XbseLf3e1YkunKRQKpd6puxWvhRRz8GrI\nZwaoHXKgdkjnvbADUI8t9W5H3Tt5CoVCoZSmrjV5CoVCofBUlSdPoVAolPpElU6e6nTSoXZIh9oh\nD7XYUu92qNLJUygUCoWHavIUCoWiAqgmT6FQKAsQVTp5qtNJh9ohHWqHPNRiS73boUonT6FQKBQe\nqslTKBSKCqCaPIVCoSxAVOnkqU4nHWqHdKgd8lCLLfVuhyqdPIVCoVB4qCZPoVAoKoBq8hQKhbIA\nUaWTpzqddKgd0qF2yEMtttS7Hap08hQKhULhoZo8hUKhqACqyVMoFMoCRJVOnup00qF2SIfaIQ+1\n2FLvdqjSyVMoFAqFh2ryFAqFogKoJk+hUCgLEFU6earTSYfaIR1qhzzUYku926FKJ0+hUCgUHqrJ\nUygUigqgmjyFQqEsQESd/PHjx/G1r30Nv/jFL7Lb+vr6cP/99+Oee+7BkSNHKt6uFFSnkw61QzrU\nDnmoxZZ6t0MvtkMikcDNN9+MkydPAgDS6TR27dqFbdu2IRwOo6enB93d3bK2r1u3DhqNRjGjKBQK\nhcKj+/rXv/71cju0trbC7/fD7XZjw4YNcLvdGBkZwebNm2E2m3H8+HEsWrQIoVBI8vauri7YbLai\nxxsYGMDy5curMmrz5s1V/b0Uli5dqvgxqB3SoXZI572wA1CPLfViRynfKRrJFxIKhcCyLHbu3AkA\nYFkWHMchlUrJ2k6hUCgU5ZE98Wq1Wv9fe/cX0tQbhwH8seVmdtKtwqZmf+wmJCfddNNF0UWg3SQW\n9IcoK4KGeCkEIUJBgTcVLjFIaF1EUVQXFtW6ijIpLYPaRbgKa1sOy2abbW7n/V3UVvJzubmjx3N6\nPjfCO3d8H16+X8a7c14RDoexZ88e7N69Gz9+/EBZWVnG4zOJ+3TpY470MUdm9JJF6znS+iT/512W\nVqsVXq8X4XAYsiwjGAzCaDRmPJ6K2WzGw4cPswpVWVmZ9TXSMdN/gzkywxzpma0cgH6yaCGH2Wye\ndHzK++Rv3bqFly9fYmRkBBUVFThy5Aj6+/vhdDoBAPv374fNZgOAjMeJiGhmzbmHoYiISDl8GIqI\nSMfY5ImIdIxNnohIxzK+T14tnZ2dGBwcxMKFC3Ho0CFYLBa8evUK169fRzgcxoEDB7Bu3ToASDnu\ndrvhdDpRUVGBffv2aTbHtWvX4Ha7UVBQgPr6+pTfqs/1HA6HA16vF0ajEZs2bcLmzZs1lyMcDqO1\ntTV5PY/Hg0uXLmkuBwB0d3fjwYMHiEQiaGhoQHFx8aznyDRLqprWWq2nmq8itS40pqenR1y9elXI\nsiyOHz8uIpGI+PLli2hubhZCCBGPxycdF0KI/v5+0dPTI5xOp1rTT5pODlmWJ1zjxYsX4sqVK2pM\nPymb9XA4HCIQCKg19QmUWI/379+L9vZ2NaafNN31iMViyUyfP38WDodDzRhCiKmzCJG6prVU60JM\nPd9sal1z2zWSJCEWi8Hn86G4uBhGoxEWiwXLli2Dz+eD3++fdBwAbDYbJElSOcFP08nh9/uT74/F\nYnj9+jWKiopUTJHdegATn8FQU7brAQB3795FdXW1Sgl+mu56CCEgyzIikQgkScK3b99UzZFOFiB1\nTWup1oG/zzfbWtfMdk3C48ePUVNTk/HxCnNNtjmOHTsGo9GIuro6FWb/23RyjI2NAQDy8vJw9uxZ\nlJWVoa6uDkuXLlUpRfbrMTo6iuHhYaxcuVKF2f823Rzz58/Hjh07cO7cOUiSBL/fj+/fv6vaKKfK\nohXZ5si21jX1Sf758+coKSlBaWnpnDpGIVNK5GhtbUVtbS3a29s1l2P58uUAgIMHD+LkyZPYuHEj\nbt68qbkcf66Hy+VS/Z/dZJujqqoKTU1NOHr0KMLhsKoNPp0sWqBEjmxrXTNNfmBgAG/evMG2bdsA\nTDxeIRQKTXqMwp/jCWpvDyiVAwCKiooQCoXUiKFoDpPJBJPJpEYMRXLE43H09fVhw4YNqmRQKkdC\nb28v1q9fr0YMAOlnSUhV01qp9YS/zTebWtfME68NDQ1YsmQJ5s2bhxUrVqC+vj7jYxQmO6JBizna\n2trw9etXmM1m7Ny5E1arVZM5Ojo6MDQ0hMWLF2Pv3r2q3CWkRI6nT5/C7/dj+/btsz7/BCVyXLhw\nAV6vF5IkwW63Iz8/f85nSVXTWqv1VPNVotY10+SJiChzmtmuISKizLHJExHpGJs8EZGOsckTEekY\nmzwRkY6xyRMR6RibPNEvmZxW2NXVhWg0OoOzIVIGmzzRLzk5OWn/7p07dxCJRGZwNkTK0NwBZURK\neffuHdra2pCXl4fKysrkY+VjY2O4ceMGhoeHMTg4iC1btqCmpgYAEI1GceLECYyMjOD06dMwGAxo\nbGxMHq7m8Xhw8eJFxGIxWCwW2O12FBQUqJaRSHPnyRMp5dSpU+LRo0dCCCHcbrfYtWtX8rVgMCiE\nECIUConDhw+LSCQy4b12u12Mjo5OGBsfHxeNjY3C5/MJIYS4f/++OH/+/ExGIJoSP8nTPykajcLj\n8aCpqQkAsHbtWuTm5iZfNxgM6O3tRSAQgNFoxMePH1FeXv7Xa3769AnBYBAdHR0AAFmW/3f4F9Fs\nY5Onf5LBYIDBYJj0tQ8fPuDMmTOorq7GqlWrUFhYmPY1c3Nz0dzcnNH+PtFM4hev9E8yGAwoLy/H\ns2fPAPw89zvxRerbt29hs9mwdetW5OfnY2ho6H/vX7RoUXJclmUAQElJCSRJwr1795JjiZ9EajG0\ntLS0qD0JIjWUlpais7MTT548wYIFCzAwMIDa2lpYrVa4XC64XC4EAgEUFhbCarVO+MfWJpMJly9f\nRl9fH+LxOFavXo2cnBxUVVWhq6sLt2/fRnd3N8bHx7FmzRoVU9K/jkcNExHpGLdriIh0jE2eiEjH\n2OSJiHSMTZ6ISMfY5ImIdIxNnohIx9jkiYh0jE2eiEjH/gN5l7Co2WRq9QAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x109553ad0>"
]
}
],
"prompt_number": 94
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Augmented Dicky-Fuller test for unit roots\n",
"# Null Hypothesis: Unit Root\n",
"# Alternative Hypothesis: AR(p), where here p is selected as the lag <= 12 that minimizes AIC\n",
"print 'Augmented Dicky-Fuller Test (p <= 12)'\n",
"print sm.tsa.adfuller(data['Unit Sales'], 12)[1], '=> we cannot reject unit root in data'\n",
"print sm.tsa.adfuller(data['Unit Sales'].diff()[1:], 11)[1], '=> we can reject unit root in differenced data at 1%'"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Augmented Dicky-Fuller Test (p <= 12)\n",
"0.453907629823 => we cannot reject unit root in data\n",
"0.00470416125858 => we can reject unit root in differenced data at 1%\n"
]
}
],
"prompt_number": 134
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# So it looks like an appropriate model might be ARIMA(12,1,0)\n",
"p=12\n",
"d=1\n",
"q=0\n",
"mod = sm.tsa.ARIMA(data['Unit Sales'], order=[p,d,q], freq='M') # mod is an object of class ARIMA\n",
"res = mod.fit(); # res is an object of class ARIMAResults"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"[ 6.19 0.04 0.18 0.19 -0.54 -0.46 -0.85 -0.52 -0.73 0.1 -0.31 -0.37\n",
" 1.09]\n"
]
}
],
"prompt_number": 135
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print res.summary() # sometimes the Display() command isn't as nice as print\n",
"\n",
"# The roots table gives you information about the stationarity of the estimated ARIMA\n",
"# model (i.e. did you choose a high enough `d`). All the roots are greater than 1 in\n",
"# modulus, so our model is stationary. We could guess this would happen based on the\n",
"# ADF test of the differenced data, which told us that we had one unit root."
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
" ARIMA Model Results \n",
"==============================================================================\n",
"Dep. Variable: D.Unit Sales No. Observations: 160\n",
"Model: ARIMA(12, 1, 0) Log Likelihood -1150.617\n",
"Method: css-mle S.D. of innovations 314.595\n",
"Date: Mon, 18 Nov 2013 AIC 2329.233\n",
"Time: 13:00:25 BIC 2372.285\n",
"Sample: 06-01-2000 HQIC 2346.715\n",
" - 09-01-2013 \n",
"=======================================================================================\n",
" coef std err z P>|z| [95.0% Conf. Int.]\n",
"---------------------------------------------------------------------------------------\n",
"const 7.6884 12.774 0.602 0.548 -17.348 32.725\n",
"ar.L1.D.Unit Sales -0.2200 0.068 -3.228 0.002 -0.354 -0.086\n",
"ar.L2.D.Unit Sales -0.1089 0.070 -1.550 0.123 -0.247 0.029\n",
"ar.L3.D.Unit Sales -0.0310 0.070 -0.440 0.660 -0.169 0.107\n",
"ar.L4.D.Unit Sales -0.1597 0.070 -2.270 0.025 -0.298 -0.022\n",
"ar.L5.D.Unit Sales -0.1137 0.070 -1.635 0.104 -0.250 0.023\n",
"ar.L6.D.Unit Sales -0.2565 0.066 -3.861 0.000 -0.387 -0.126\n",
"ar.L7.D.Unit Sales -0.2117 0.068 -3.114 0.002 -0.345 -0.078\n",
"ar.L8.D.Unit Sales -0.2040 0.069 -2.942 0.004 -0.340 -0.068\n",
"ar.L9.D.Unit Sales -0.0047 0.070 -0.067 0.947 -0.143 0.133\n",
"ar.L10.D.Unit Sales -0.0956 0.071 -1.355 0.177 -0.234 0.043\n",
"ar.L11.D.Unit Sales -0.0327 0.070 -0.465 0.642 -0.170 0.105\n",
"ar.L12.D.Unit Sales 0.4781 0.068 7.082 0.000 0.346 0.610\n",
" Roots \n",
"==============================================================================\n",
" Real Imaginary Modulus Frequency\n",
"------------------------------------------------------------------------------\n",
"AR.1 -1.1017 -0.0000j 1.1017 -0.5000\n",
"AR.2 -0.9003 -0.5025j 1.0311 -0.4190\n",
"AR.3 -0.9003 +0.5025j 1.0311 0.4190\n",
"AR.4 -0.5342 -0.8814j 1.0307 -0.3367\n",
"AR.5 -0.5342 +0.8814j 1.0307 0.3367\n",
"AR.6 0.0008 -1.0515j 1.0515 -0.2499\n",
"AR.7 0.0008 +1.0515j 1.0515 0.2499\n",
"AR.8 0.5202 -0.9593j 1.0913 -0.1709\n",
"AR.9 0.5202 +0.9593j 1.0913 0.1709\n",
"AR.10 0.8766 -0.5080j 1.0131 -0.0836\n",
"AR.11 0.8766 +0.5080j 1.0131 0.0836\n",
"AR.12 1.2439 -0.0000j 1.2439 -0.0000\n",
"------------------------------------------------------------------------------\n"
]
}
],
"prompt_number": 136
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"predict = res.predict('2012-1-01','2015-01-01', dynamic=True)\n",
"\n",
"fig = plt.figure(figsize=(10,3))\n",
"ax = fig.add_subplot(111)\n",
"\n",
"ax.plot(data.index, data['Unit Sales']-data['Unit Sales'].mean())\n",
"ax.plot(predict.index, predict, 'r-');"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAl8AAADJCAYAAAAdH1dqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsnXeYW9W19l9p1EejMr27G2OPbVzoxaa3BEjAIUCIgZi0\nLwGcS3JvYsABh3vjBANOMCUh2JhAEoZgSug2I/fexuM2vVdp1HvZ3x9npJFmJI2OdFRmvH/Pkyec\nrVO21+gcrbPW2u/iEUIIKBQKhUKhUCgpgZ/uCVAoFAqFQqGcS1Dni0KhUCgUCiWFUOeLQqFQKBQK\nJYVQ54tCoVAoFAolhVDni0KhUCgUCiWFUOeLQqFQKBQKJYUI4j3wjTfeQEdHB7Kzs/GDH/wAarUa\nGzZsQHd3N0QiEZYsWYKlS5cCAGpra/Hee+/BZrPhgQceQFVVVdRxCoVCoVAolIlK3M7XQw89BAA4\ncOAAvvrqK3znO98Bj8fDypUrkZ+fH9jP5/OhuroaTz75JKxWK1588UVUVVWFHZ8zZw54PF7i/yoK\nhUKhUCiUDCVu58uPXC6H1+sNbI/UbO3t7UVJSQlEIhFEIhGKiorQ09MDQsiocf++FAqFQqFQKBOV\nhJ2v3bt345ZbbgEASCQSrF+/HhUVFbjzzjuRn58Pi8UCmUyGTZs2AQBkMhlsNhu8Xm/YcQqFQqFQ\nKJSJTELO16FDh1BaWoqysjIAw6nIuro6bNmyBQ8//DDkcjlsNhtWrFgBQgg2btyIiooKaLXasOOR\n2LZtWyJTpVAoFAqFQkkZKpUKixYtCvtZ3M5XU1MTTp06he9///ujPhOLxRCLxQCA4uJidHd3w2az\nwefzwWQyQSQSRRyPxrXXXhvvdCkUCoVCoVBSRrSgUdzO1wsvvIC8vDw8/fTTqKysxIMPPohXX30V\nAwMDyM3NxX333QcA4PP5WLZsGdasWQMAWL58edRxCoVCoVAolIkMj4yskM9Qtm3blvTIl0ajCchj\nUMaG2os91GbsoPZiB7UXO6i92EHtxY5ofgsVWaVQKBQKhUJJITTyRaFQKBQKhcIxNPJFoVCSwule\nE375wYl0T4NCoVDGFdT5CkKj0aR7CuMKai/2TDSb7W7W4WSPMWnnn2j2SjbUXuyg9mIHtRd3UOeL\nQqHEzeEOA7QWF9xeX7qnQqFQKOMGWvNFoVDiwuXx4bqXdkIs4GPj9xajXCVN95QoFAolY6A1XxTK\nOOdIhx4/+deRdE8jhLoeIybnyTAlLxt9Jke6p0OhUCjjBup8BUHz2eyg9mJPvDbb2zKIox1GWF0e\nbieUAIc7DFhcoUaxQoyeJDlf9DvGDmovdlB7sYPaizuo80WhjAMOd+ghEfJxvCt5xe1sOdKhx6JK\nFUoUkqQ5XxQKhTIRoc5XEFS5lx3UXuyJx2Y2lweNA1Z8a34ZDrfruZ9UHDg9XpzqMWN+mQrFCknS\n0o70O8YOai92UHuxg9qLO6jzRaFkOLVdRswqkuPyqXk40mFI93QAAHU9JkzJz4ZcLEBxhkS+CCHw\njY/1QxQK5RwnbufrjTfewNNPP43nnnsOej3zNl5bW4unnnoKjz/+OOrq6gL7sh1PFzSfzQ5qL/bE\nY7PDHQYsrFCjqkSBJq0Vtgyo+zrSbsCiChUAoEQhQW8G1Hy9f7wbS9fvwE/+dRSv7mrG2T5zUuaU\nydB7kh3UXuyg9uKOuJ2vhx56CKtXr8ZVV12Fr776CoQQVFdX44knnsCqVatQXV0NAPD5fDGPjxPV\nCwolpRzu0GNRhQoSYRZmFeVkRN3X4Q49FlaoAQBFORL0mZ1pv39bB624Z3EF7ltcAZfHh0feO04j\nYRQKJSNJOO0ol8vh8XjQ09ODkpISiEQiqNVqFBUVoaenB729vTGP9/b2cvFvihuaz2YHtRd72NrM\nX+81t1QJAFhUocLhNKcenR4vTvWacUE5MyepKAtSYRYGbe6Q/fa06ODxJSa+ysZe/WYnpuVn44pp\n+Xhk6XQopUKc6jEldP3xBr0n2UHtxQ5qL+4QJHqC3bt345ZbboHFYoFMJsOmTZsAADKZDDabDV6v\nl9U4hUIZxl/vJRFmAQAWVarx8o6mtM6prseEafnZyBYNPz6Kh1KPedkiAIDb68OvPjiBN+9fjGn5\n8pTMq8/sRFGOJLB91fR87GjSomrIcaVQKJRMISHn69ChQygtLUVZWRm6u7ths9mwYsUKEEKwceNG\nVFRUQKvVshqPhkajCXje/twzl9vHjh3DY489lrTzT7Rtai/22/6xWPc/ya/Awgp1YPuSy69Eo9aK\nL7bVQJzFw5VXLcEnJ3sgGzgLAZ+Xkn9PXbcJap8p5H4UOM3Ytvcw5nz7agDAPz/bDqeHwDAUDUuF\nvTq0PhTliAPbcgvB5wMy/PTKaRnz98+079e5vu0fy5T5ZPq2fyxT5pPp21EhcdLY2EjefPPNwLbX\n6yWrVq0ier2e6HQ6snbt2rjGI7F169Z4pxozNTU1Sb/GRILaiz1sbfbg3w+SA626kLGH3zlM9jRr\nidPtJf/9QS25+LmvySd13RzOMjp/3HqWvH2wLWTsuW1nyVsHhsf+ebidLP7DNrL1TF9C14rVXm6P\nl1zy3NfE7fUGxjxeH7n+pR2kU29LaA7jCXpPsoPaix3UXuyI5rfEHfl64YUXkJeXh6effhqVlZV4\n8MEHsWzZMqxZswYAsHz5cgAAn89nNZ5O/F4rJTaovdjDxmYj6738LKpQYVeTDn8/2I5skQC/+8Yc\nvH2wHbfMKeF4tuHRWV2YN2JOJQoJuo3DKx5PdJsgE2ZBb3MldK1Y7aW1upArE0HA5wfGsvg8XDE1\nHzubtPjuouhR9YkCvSfZQe3FDmov7ojb+XrppZdGjc2fPx/r1q1LeJxCoYyu9/KzqFKNn/zrKG6f\nV4JfXz8LAPAnTSNO95pwfrEi6fPSWZ2B2i4/xQpJiAbZiW4jLpmSC73dPfLwpNBncqBwKOUYzFXT\n8/Hukc5zxvmiUCjjA/7Yu5w7xJSnpQSg9mIPG5vtb9Nj0ZCcQzAXlCvxh9vnYtUNs5DF5yGLz8O3\nLyhF9dEuDmcaGZ3VhTz5aOfLr/Wls7pgcXowv0wZqPmKl1jtxRTbj3a+Lp6Ui1O9JpgdqXEC0w29\nJ9lB7cUOai/uoM4XhZKBEEKgaRjA0hkFoz4T8Pm4emYBeDxeYOz2uaWoaRiAIQWRJq3FhfzsUEcn\nWGi1rtuIOSUKqGUiGOyJpR1jZeRKRz9SURYWlKuwt2UwJfOgUCiUWKDOVxA0n80Oai/2xGqzxgEr\nfIRgZmFsMg1qmQhXTsvDxyd6Epjd2NhcHngJQbYoNBWqkgrh8Phgc3lQ223E3FIl1FIh9AlGvmK1\nV585fNoRAK4ckpw4F6D3JDuovdhB7cUd1PmiUDKQmoZ+LJ0RGt0ai2ULyvHvY51JVXUftLqQly0a\nNS8ejzeUenTiRLcJcwORr9Sk+/rNThQpRke+AOCqafnY16LD3Rv34+nPTuH9Y10Ji79SKBRKIlDn\nKwiaz2YHtRd7YrVZTf0ArgmTcoxGVYkCORIhDrXr45hZbGiHnK9wFOeI0Wmw40yfGVWlSiilwoRX\nO8Ze8+VAoTx85CtfLsZnP70CT98yG3NLlXh1dzNadRNT0Jnek+yg9mIHtRd3UOeLQskwOvQ2DNrc\nmFvGTpmdx+NhfpkSDf2WJM2MKaYfWe/lp1gpwa4mLYoVEsjFAqhlQhjs7pT0fOwzO1GkCD8vABBm\n8TGrKAffnl+GCpUMZmf6m5NTKJRzF+p8BUHz2eyg9mJPLDZjCu3zwWeRcvRTrpKiy2iPY2axoQ0j\nM+GnOEeCrxsGMK+UkbsQC7IgyuLD6vLGfb1Y7OX2+mC0uyPOayQKiQBmx8R0vug9yQ5qL3ZQe3EH\ndb4olCAIIbC50vvDXFM/gKtZphz9lKmk6DQkz/nSRUk7ligkMNrdIb0UVbLEU49jMWBxIjc7VGA1\nGnKx4JyRnqBQKJkJdb6CoPlsdkxEe2katPjOG/thSVJaaiyb9ZudaNPbsKhytL5XLJSrpOgyOMbe\nMU50VhfyI9RWFSuZgve5pcNCryqpMKGi+1i+Y/0RNL4ikSMRTNi040S8J5MJtRc7qL24gzpfFEoQ\nu5q18PgIXtrRlJbraxoGcPnUPAiz4rs1/XpbXl/idVYne0zweENXBWotkSNfZUoplFIhpuRlB8bU\nMlHCchNj0Wd2hNX4ikSORDhhnS8KhTI+iLu90OnTp7F582bMnj0b999/PwBgw4YN6O7uhkgkwpIl\nSwL54draWrz33nuw2Wx44IEHUFVVFXU8XdB8Njsmmr0IIdjXMoh135qHX35QixvPL8KCchWn1xjL\nZruatbhtbmnc55cIs6CSCjFgcaI4gvRCrPz201P45XUzcdGk3MCYzupEfqSaL4UE76+4JKRWjYl8\nxZ92jOU71mdyRtT4CkeOWIB+szPuOWUyE+2eTDbUXuyg9uKOuCNfbrcb3/rWt0LGeDweVq5cidWr\nVwf+SD6fD9XV1XjiiSewatUqVFdXRxxPxaooCiUSTVorBFk8zC7OwS+vPQ+/+/wMnJ74i8XZ4vL4\ncLzTiAvjTDn6KVNJ0MVB3ZfB7kaz1hoyxkhNRHZ0FBJhyHYyIl/HOw1wuIf/Lv0WdmlHhUQAs5PW\nfFEolPQRt/M1b948yOWj1bdHOlC9vb0oKSmBSCSCWq1GUVERenp6wo739vbGOx1OoPlsdkw0e+1r\nHcSlk/PA4/Fw9cwCzCjIxut7Wjm9RjSb1XYbMTlPBqVUGHGfWChVJl507yMEJocbTUHOl9dHYLC7\nkSuLfX7JqPn6/Vdn8enJ4WdFn4ld2pEpuJ+YaceJdk8mG2ovdlB7cUfcacdwSCQSrF+/HhUVFbjz\nzjuRn58Pi8UCmUyGTZs2AQBkMhlsNhu8Xm/YcQolXext0WHZgvLA9uPXzsRdf9uHBy+ZBJmI01sl\nLPtbB3Hx5NyxdxwDLuQmzA4PfAQhkS+D3Q2FRAABi3o0tUyIFp117B3ZzM3pwQcnuvHtC8oAAH2s\nI1+05otCoaQXTgvuH3roIfzud7/D5Zdfji1btgAA5HI5bDYb7r33Xtxzzz1wOByoqKiIOJ5OaD6b\nHRPJXnaXF3XdJiwOSvnly8WYX6bCTg77Akaz2YG2QVw8KXHniwu5Cb+j1ayzBqLZWosTebLYtLT8\nqKSJtRgKZy+L04MeoyMgJsvUfLEouJ/Aka+JdE+mAmovdlB7cUdCzlekGi2xWAyxmHkTLS4uRnd3\nN2w2G6xWK0wmE0QiUcTxaASHPDUaDd2m25xtb/p0B4olXsjFgpDPb5hViC/P9Cf9+p9urUFTvxlz\nhzSyEjlfuUqKMx39Cc2nZs9+KLM8EGXx0W9xQqPRoGbfoYDMRKznUw21GOLKXh6fD3a3F4tUbrz8\n2QG4PD6YHG7UHdoT8/nkYgEGjJak/j3pNt2m23Q7KiROtmzZQlavXk0effRR8tprrxFCCHnllVfI\nM888Q1566SWi1+sD+x47doz84he/IL/4xS/I8ePHxxwPx9atW+OdaszU1NQk/RoTiYlkrz9uPUve\n2NsyatzscJMlL2qI0e7i5DqRbLb1TB95pPoYJ9fQWZzk2j/vSOgcOxoHyCPVx8hP/nmE7GnWEkII\n+bC2i6z+5CSr87QPWsltr+2Oex4j7WWwucjV67eTTr2NXPvnHaRZayG3vrKL1TlNdhdZ8qIm7jll\nMhPpnkwF1F7soPZiRzS/Je5CljvuuAN33HFHyNiPf/zjsPvOnz8f69ati3mcQkk1e1sG8ew354wa\nl4sFuHBSLjQNAwlJQIwFU++V2CpHP2qZEG6PDxanJxDJY4vB7oZKKkSORIAmrRWXTsmLqm4feS4i\nGDhc7Wh2epAjEaBMJcXMAjnePdLJqtgeALLFAtjdXnh9BFl89i2cKBQKJVGoyGoQNJ/Njoliry6D\nHRanGzMLR6/eBYAbZhXiqzP9nFwrks32tw2G6GklAo/HQ5lKmpDchNHuhlIqxLT87EDRvW4MmYlw\nZIuy4PL64pbsGGkvi9ODnCGH8o55pfjwRHfUhtrh4PN4yBYJktbFIJ1MlHsyVVB7sYPaizuo80U5\n5/GvMozUyPrKafmo6zFh0JqcHoWdehtcHh+m5WePvXOMlCklCRXdM86XAFPz5WjSMoXtOqsrosBq\nJHg8XsJyE8GYHW7kSBjna8mMfEiFWSiM0O4oGhO5xRCFQsl8qPMVxJgFcpQQkmGvPrMDj/37eEoF\nd8/0mVFVooz4uUSYhcun5mFbfeLRr3A229+mx8WTc8GL4PzFQ6KRLybtKMLUPBladDb4CIHO6kSe\nnJ3zBSSWehxpL7PTA7mY0RkTC7Lw3YUVmF2sCHNkdHImaHNt+gxjB7UXO6i9uIM6X5SM4oWvG7C7\nWQddkqJM4WgdtGJKnizqPjfMKsJXZ/qScv39rdylHP2UJaj15U875kiYuq8eoyNqX8do+Fc8ckFw\n2hEAHr58Cm44v4j1eWjki0KhpBPqfAVB89ns4Npee1p0ONNnxsxCecI6VWxo0dkwOS96yu+Sybmo\n77fAmGD6LJzNGgYsmF2Sk9B5R1KeoNaXv+AeAKbmZaNZZx1KO7JP8allQujjtNtIe5kcoc5XvMjF\nwgmp9UWfYeyg9mIHtRd3UOeLkhE43F78cWs9fnndeZiWn50y58tgc8Ht9Y1ZyyQS8FGYI8GAhfuG\nzCYHk+LjkoQL7h1uKIf6NE7Lz0ZdjwleQpAtymJ9rkSFVoMxOz2QSxJ3vhQ08kWhUNIIdb6CoPls\ndnBpr80H2jCzUI7Lp+ahXCVFR4qcr9ZBG6bkZcdUb5UrSzx9NtJmXh8ZlUrjghKFBP0WJzxeX1zH\n+9OOADA1PxsHWweRly2Kqy5NJRPCEKfdRtrLwlnki9Z8Uai92ELtxR3U+aKknW6jHe8e7cIvrpkB\nAChXydCpT02fz9ZBGyblRq/38qOWiaDnULMKYCI52SIB53pTwiw+8rPF6DWzj9QRQgKrHQFgWr4c\np3rNcaUcAUAtFXJmN7PTzYnzRSNfFAolnVDnKwiaz2YHV/Y63GHAJZNzA2KZFerEexPGSovOiilj\n1Hv5UcuEGEww8jW6hmk4wsQ1ZSpJXKlHq8sLYRYfYgGTYpySJ4OXkLiK7YEhp5Wjmi+u0o45tOaL\nAmovtlB7cQd1vihpp1lrDdG4SrRYnA2tOhsmj7HS0U8iTkQkTEMNrJNBmUqKjjgiiMHF9gAgEwlQ\nqpQgPw6ZCYBZ7Wi0c7TakaO0I13tSKFQ0gl1voKg+Wx2cGWvFp0VU4OiTyqpEF4fSXhlYSy0Dlox\nOTe2yFeuTAR9ghIYI21mdLihkCQn8rWoQo2ahgHWxwXXe/mZmp8dd+RLJYs/7RhO54sLezE6XxPP\n+aLPMHZQe7GD2os74n6FPH36NDZv3ozZs2fj/vvvBwDU1tbivffeg81mwwMPPICqqqq4xinnFs1a\nK6YERb54PB4q1DJ0GuxJS8kBzApLndWFMlVsvQHVHOpV+THaPUn7N153XiHWaxrRrLViKgv1fGal\nY+ij4Y55pQnUfHEXMTQn0K8ymByJcEIW3FMolPFB3E8xt9uNb33rWzh79iwAwOfzobq6Gk8++SSs\nVitefPFFVFVVsRqfM2cOpyrfbKH5bHZwYS+by4NBmwtlSmnIOJN6tGFOCXv18lhpG7ShTCWFgB9b\nAFidLcJggoXjYWu+khT5Embxcef8Mrx7pBP/c8N5MR9nsLmhkoVGuZZML4h7HgqpABaHJ65G1qN6\nOzo8gfZCiZAjnphpR/oMYwe1Fzuovbgj7rTjvHnzIJcPNyLu7e1FSUkJRCIR1Go1ioqK0NPTw2q8\nt7eXk38UZfzQorOhUi0b9aNcrpKiQ5/cuq+2QRsmx7jSEWAK7g0c1S75MTmSV/MFAN+aX4ovz/TB\nxCLKY+TYIRTw+ZCLs1jNIRwenw9Ojw8yIXutsZHQmi8KhZJOOKv5slgskMlk2LRpEzZt2gSZTAab\nzcZ6PJ3QfDY7uLBXiy58SiwVKx5bdNYxle2DyZUlHvkaVfNl90CRxNRqvlyMK6bl4cPanpiPGVlw\nzwWqOGU6gu1lcXohF2dxEh2nNV8UgNqLLdRe3MHZK7dcLofNZsOKFStACMHGjRtRUVEBrVbLajwa\nGo0mEPb0fwm43D527FhSzz/RtrmwVzPKMTUve9TnAy1ncbKLAJidtPkfbPThrstiPz8hBHY34PL4\nsGfXjriu78e/bXQUYI5EkdS/190LK7Dy3UMoszbhmquvHnN/o90N72A3NJo2zubDd9lQs+cApt42\n9vWDt/1oNBpoHQRysZiT+ezdtQNeH4HT44VYkJUR9xMX234yZT6Zvu0nU+aT6dt+MmU+mb4dFZIA\ndXV1ZPPmzYQQQrxeL1m1ahXR6/VEp9ORtWvXxjUeia1btyYyVUqG8th7x0hNff+o8T6Tg9zw0s6k\nXvu7b+wjp3tNrI65acNO0muyczaHn1cfJbuaBjg7XyQefOtgWDuH438+PEG+ONXL6fUff/842bCj\nkdhdnrjPcbLHSL735gHO5nTDSzvJgNnB2fkoFAolmGh+S9yRrw8++ADHjh2DwWCA3W7HD3/4Qyxb\ntgxr1qwBACxfvhwAwOfzWY1Tzi2aI4ic5stFsLo8sLk8kIm4r4ny+gg6DHZMUsde8wUwWl8Gmzsg\nCJsoRnvypCaC+d6FlVj71VnorC7cNrcEwqzIFQeGMFITifLAJZPxl90tuPXV3bhmZiF+cOlkFCvY\n2dDs4Galox9/3Ve+PL5VnBQKhRIvcT/J7rjjDtxxxx0hY/Pnz8e6detG7ct2PF1oNMNpTcrYJGov\nuyuy1AOfx0PZkNjqzMKcBGYZnh6jHbkyEaQsG0XnJqhyP9JmJkfypCaCuea8QhQpJHh1VzPe3N+G\nlVfPwNUzw69gDKfzlShzShRYf9d89Jud2LCjCZsPtOFX1429AjPYXlz3wFRMwLov+gxjB7UXO6i9\nuCPy6y+FkmRaBq2oVMsiSj1UqKToTNKKxxYWPR2DYfo7crfi0ZREkdWRzClR4M/LLsDj187Eek1D\nxP2MDu4L7v0U5ohx5wVlONFtYn2s2cmNzIQfuUQAs5NqfVEolNRDna8gqEfPjkTt1TKG+Gcy2wy1\nsujpGIw6wRWPwTbz+gisTi+n0ZxYuGiSGgMWFwghYT832pOnPQYAs4py0Dpohc01dtQp2F6cpx0n\nYOSLPsPYQe3FDmov7qDOFyVtMPVekaNP5SopOjh2vkwON9490on3jnVhVpF87ANGoJZxp3Jvdnog\nE2WxFh5NFIkwC1IhH4YwqvMOt3don+Q9GkQCPmYUyHGq18zqOIuTW020HIkQpgnmfFEolPEBdb6C\niGl5KCVAovZq1lkxNS+yA8S0GOJO++3lnU24/S97cazLgN/cMAs3zS5mfY7cOPWq/ATbzOTgvrYq\nVgpzJOg3O0eNG4aiXsnuNDGvVIkT3cYx9wu2l9npQY6YO3vliAWwTDChVfoMYwe1FzuovbgjtfkO\nyjmNweaCWJAVKHJv0VoxJT965IvLtONHJ3qw8b5FrIRVR8JlzZfJnlx1+2gU5ojRZ3bivKLQxQzJ\nKLYPx7xSJT6ui134FeA+7aiQCBIWzY0VQkhaW6dRKJTMgka+gqD5bHawtdczn5/B9zYfQEO/BQ63\nF1qrC+UqacT9ixRiDFrdcHq8Cc6U+fEzOdys5Q1Gok5wtWOwzYwOT8qK7UdSKBej3+wYNZ4MmYlw\nzC1jIl+R6s78BNvLwnXBvViQkubau5u1eOzftUm/DkCfYWyh9mIHtRd3UOeLkhI8Ph+OdOhx14Jy\n/PTdo3hlVzMqxmhqLeDzUawQo8sw2klgi93tBZ/HgyTBvoBqmShsrVQ8pCrKFI7CHDH6LaPTjsYk\ntBYKR4FcDJlIgDZ97GllM9dSExJhSvo71vdbsK9Vx+kqWQqFMr6hzlcQNJ/NDjb2OtNrRolCgnsW\nVeC17y7E3hYdpheMXfA+szAHZ/rYFWaHgys9LS50vobnlNxVhdEozBFHrvlKkUM4r0yJ2q7hui+t\nxYl/HekI2SfUXtymHeViQUoK7tv1Ngiz+KhpGEj6tegzjB3UXuyg9uIO6nxRUsLBdj0WT1IDAKbm\nZ2Pz/RfiV9fNHPO42cUKnOxhrwk1Eq7kE6TCLBDCCMQmCqPxlZ6ar6IcCfrCOF+pjMbNLVWE6H29\nsqsZb+xti7i/xenmNO2YI0lNwX37oB13zCvFtrP9Sb8WhUIZH1DnKwiaz2YHG3sdbNPjwsrcwLZE\nmIWcGJyhqlIF6nrGXhU3FkxEJ/Efbh6Px8hN2OOLfoXUfNk9UKQz7RjO+UphNG5eqRK1QyseGwcs\n2NWkhdHhhsfnC+wTovOVBIV7UwpqvjoMNnx3UQVO9piSnnqkzzB2UHuxg429HG4v/nm4Y+wdz1Go\n80VJOk6PF3U9JiysULE+9vyiHDTrrAkX3XOZ4lMnKDfhJ5WOzkiYmi/HqIJ3Q4pqvgBgRoEcvUYH\nzA43/rS9EQ9eMhlqqRCD1tG2dXt9cHsIpAnW7AUjlwiTHvkyOdxwun0oU0pw6ZS8lKQeKZRM4HSf\nGeu+bggpLfBzqF2PvjALfs4lOHe+NmzYgFWrVuHpp5/G9u3bAQC1tbV46qmn8Pjjj6Ouri6wb6Tx\ndEHz2eyI1V513SZMzc+Oq15HIsxCpVqGhn4L62OD4bKBda5MFHfd12idr/SkHbNFAgj4/FEF53qb\nCypZapwvQRYfs4pz8Le9rejU23HnBWXIl4sxELQQwG8vi9MDuUTAqVyDXJwFq9ML3xgrLhOhQ29H\nZa4MPB4P151XmPTUI32GsYPaix1s7FXfb4ZCIsDbh9pDxrUWJ371wQl8782DePdIJ7y+5N1/mQzn\nzhePx8PKlSuxevVqLFmyBD6fD9XV1XjiiSewatUqVFdXA0DY8bGWnVPGJwfb9VhcqY77+KoSBeoS\nrPvisoE1Vyr3pjRKTQDhU48tOiumxNHzMl7mlSrx9qEO/OyqaRBm8ZEvF0FrHZ0O5TrlCDCraSVC\nPmwc1O8vFOUNAAAgAElEQVRFom3Qhko1I6dy+dQ8nOwxwTD03ek1ObDu63q4vb5op6BQxiX1/RY8\neMlkHO4whOg1btzXhm9WleAv9yzEV2f6sOKdw+hKUhu5TCYpacdgJ6q3txclJSUQiURQq9UoKipC\nT09P2PHe3t5kTCdmaP6fHbHa61C7Hhcm4HzN4cD54jLypZbGn3YMrflKn9QEwGh9BRfdG2wu2Fze\nhLXQ2HDZ1DxcMTUPV88sAADkZ4uhtQw7tn57mR3canz5yZEkV+urQ29DhZpxZiXCrEDqcWeTFsvf\nOohPT/aybrMUDfoMYwe1FzvY2Ku+34ILypS4fW5JoParx2jHF6d78cAlkzAlLxuv3bMQMwvl2FLb\nnaQZZy6cP80kEgnWr1+PiooK3HnnnbBYLJDJZNi0aRMAQCaTwWazwev1hh2nTCxsLg/q+y2YX6aM\n+xxVJUps3Bd5FVwsGB1uTCuIX9k+GLUsfHSGLcxqx/Q5X0U5oUKrjVorZhTIU6rEvqBchQXlw7WA\nBXJRSNrRjyUJkS8AyBEz/R1L4v96RqVdb8PlU/MD29eeV4g/bD0LYRYfa2+fi5r6ARzp0Cd0f1DO\nQQgB3G5AJEr3TMLi8frQorNieoEchTkS3LNpP354+RT8dU8r7lxQDrWMmTefx8MF5Srsbtalecap\nh/On2UMPPQQAqKurw5YtW3DrrbfCZrNhxYoVIIRg48aNqKiogFarDTseDY1GE/C8/blnLrePHTuG\nxx57LGnnn2jbsdhLWDkX5xfnYN/unXFfb3KeDFqzHZ98VYNbr786rvk2d/aiwNkHzC1N+N+fKxNi\nV10jNLwu1sf7x76uqYHFQQIORTr+fjadD/1ycWB7Zy/B9ILE7ZPIdr5qBk71mUbZa9+R43CYhiPq\nXF0vR6KAxelJ2r+nXZ+Ne3Nlge0rLr8SZ+aWYoqzHYbGY1hYMQfVRzsxxdHKyfX8Y5nwfBgP2/6x\nTJlPrNuN/+//Qd7YiOIvv0zp9f1jY+1f/cV2KAUEEmEWJMIszMj24H/+sRONNiH+/YNLQvYvVUhw\npr0PGs1AxtiXS3tFhCSJ+vp68uabbxKv10tWrVpF9Ho90el0ZO3atYQQEnE8Elu3bk3WVAPU1NQk\n/RoTiVjs9WJNA/nL7uaEr/Xjfx4hu5oG4j7+wb8fJEc79AnPgxBCdjdryc/ePRrXsX6bGWwucvX6\n7ZzMJ162HO8iT396KrD9zGenyL+PdqZxRoRsbxggj753LLDtt9f7xzrJms9PRTgqfv7r/eOkpr6f\n8/MSQojP5yNLXtQQo90VcR+DzUWWvKghbo+Xk2vSZxg7xqW9nE5CSksJOf/8lF86Vnt9UtdNfvPR\nicD22T4TWfyHbWTjvpZR+/aZHOSmDTs5mmFmEc1v4Tzy9dprr6G/vx+5ubm47777wOfzsWzZMqxZ\nswYAsHz5cgCIOJ5O/F4rJTZisVdtlxE/uXJqwteqKlGgrtsUksJhA6cF99L4C+79NjM60lvvBfgL\n7ofTjg0DFtw+FBlMF/lyEbRBaUe/vZi0I/f2KlNJ8a8jnZhTokDBUBSQK3RWF4RZ/KipZaVUiDKV\nFKd6zZjHQeqRPsPYMS7t9e67wNSpwKFDgMuV0tRjrPY622/BzMKcwPbMwhz89ubzce15haP2zZeL\nYHJ44PR4IRZwJyWT6XDufP3oRz8aNTZ//nysW7cu5nHKxKHDYMNkDlbPVZUo8O/jXXEfz2nBPQc6\nXyZ7+tTt/RTKh/s7en0ELTorZ3Vx8VIgF2PAMtqxNTmSU/P186umYeO+NnzvzQN4/NqZuH5WEWfn\nbtfbUKke+7u/sEKFwx16TpyvicDuZi1Wf3IKt80txV0LylCqlKZ7SpkDIcDzzwO/+x3wi18A9fVA\nVVW6ZzWK+n4Lvn9RZcjYrVUlYffl83goyhGjx+TA5Nz0Pn9SCT/dE8gkYsrTUgKMZS+L0wO724u8\n7MTfzOaUKHCqxxSXHImPEJgdHig5cnbUQ/0d45mL32bGNMtMAP6Ce8b56jDYkCcTIVuUXodQLROG\nqNz77WV2ctvX0Y8gi4+HL5+C5789H6/uasG7Rzo5O3e73o7K3LEdh0UVahzpMHByzYnwDPvkZC/u\nvKAMXkLw/c0H8dtPTyXtWuPOXhoNYLcDN90EzJkDnDw5ep/+5GnJxWIvQgga+s0hka+xKFFK0GM8\nt0RXqfNFSRqdBjsqVDJOVs/ly8WQirLQEYcejNXpgUTIhyCLm6+7WJAFsYCfkDp6umUmAKaxtI8w\nTnLjALPSMd0I+HyowqjcW5zJkZrwM6dEgUeXToeGQwX6jhgjXwvKVTjRbYSH6n3B4fZiT7MO311U\ngZVXz8BHP7oMn5/qC2k5dU7z/PPAypUAnx/e+RoYAEpLgV270jO/7m7o9h6CIIs/+qW7M/KLDXW+\nznHGZf4/jYxlrw69DeUq7lIG5xcpcKaPvSaSkcN6Lz9qmQh6O/vUo99mXLY7ihcejxcQWm0YsGB6\nBjhfgD/1yETk/PbS21xJjxQuKFfiZI+JM9HT9sHYnC+lVIhSpRSn4/huj2S8P8N2NetQVaIISBHI\nRALIxVmwOJLTBmpc2evsWeDAAeD++5ntcM7X7t2AWg385CeMFAXHRLXXl18CF1wA8Y8exszCEc+S\n7m6gshKI0MmmVCFFt4k6XxQKJ3Qa7KhQc+d8lauk6Dayj3wZ7dw7OmqZEHpr/Cr3jMZXelN8wFDd\nl9mBxgFLRkS+ACA/e7SOWqvOhil5yVXez5EIUa6W4jRHoqdtentMzhfA1H1xlXocz3x1pm9U3Z1C\nIoQhBQ3QM56//hVYsQKQDj1T58wZ7czs3Ak8+ihQVga8+GJq5uX1AqtXAw8+CLzzDkRtLbgga8Rz\n+qOPmIUBr7wS9hRM5CsNKveEAC+9BHiS2+M1HNT5CmLc5f/TzFj26tDbOY18lSol6DKwfzsyJsHR\nKVdJ8aN/HcUVL2iwdP32gIJzMAMWJz4codzst5nJ4YEizWlHwN9g24nGAQumj3xbTRP58mGVe41G\nA5PDDYvTkxLl/YUVKhzp1Cd8Hq+PoNsY+8vHogo1Dnckft3x/AyzujzY3zqIpTMKQsaVUiFM9uT8\nOI4re9XXA4sXD2/PnAm0tQGOoGfirl3AlVcCGzYAa9cyn3NIWHs98wzw9dfA4cPAddfh9PzLcPHJ\nfaH7fPgh8H//B7zzDmAe/XJTqkhT2vGLL4Cf/xz4z39SfmnqfJ2j+AhBs9aa1Gt0GoZbq3BBWZyR\nL1MS6qt+e8tsbH/0Knz1/67E6/cuwut7WwM9+/y8tL0Jr+5qDnu8IQnRuHgoyhGjWWuF3uZGWYas\nKsvPDpWbaNJaMTU/G/wUKO8vKOcmAtVndkApFUIijG3p/IJyJWq7jOd0bdOuJi3ml6lG3asKCbMI\n45ynowMIFiIXiRjJibNnmW2rlYmEXXQRMG0a8NhjwCOPJH9eBw8Cv/oVUFwMANDMuBCT9349/LnZ\nzKRDf/AD4JprgL//fdQpSpRJSDv+4x/MwoS77gIeeIDZHsnzzwM33gi8+iq3144B6nwFMa7y/1F4\n4esGbDsbecWLxelhuspvPpBQR/mxa764jnxJ0R3m7ahJa8EHUXqDJUNTi8/jQSzIglSUhekFctww\nqxB/29ca+PxMnxn72wZhdnpgdQ2/tYfUfEkzIO2YI8He1kFMzc9GFj91bYWiUSAXY8A6XPPVNGBJ\nmQTGwnJV3E7QziYt/nWkAyd7TGjSWmNOOQKASiZCiUKScMqTi2eYjxB8UNud0LMhHr4604/rZ43W\ngVJKBTDFUV8ZC+PqmT/S+QJC67727wfmzx9OS/7yl0BtLROR4oiw9mpsBKZPB8D8tmydugiyXTsA\n59AL1BdfAJddBigUwE9/Crz8MpPuCyJfLoLZ4YHDzWGT+3feAS65BPjud5lo4M9/DjQ0DH9+4gTj\nrL77LmOj5vAvysmCOl8TkBM9Rnx+KnyT8hadFQ/8/RDy5WLkiIVh++hxgd3lhdnpQWEOd8KVJQoJ\n+syOUT8K+1oG8cXpvojHmezJl3VYcekUfHaqD50GOwghWK9pxMOXTUGlWoYO/ehoXY/RgUJ56hpY\nR6JwKPI1PT9z9HXy5CLogrS+mrRWTMtPTUpUJROhSCHB2T4L62Pf3N+GfS2DePaLM/jllhOYmsfO\nplfNKMDWKC9NqeKfhzvx7Bdn0DqY3Mh4MBanB4fa9aNSjgCNfAEAbDYmglQwwj7Bzpc/5ehHLAbu\nuYdxLgC8f7wL2xu5W80LgKmVam9nInBgXjoLJ5eBN2cOsH07s8+HHwK338789zXXMAsBRqzG9Gt9\n9Zo5jH4dPQosX85Evn7wA+C//5uJBPodvxdeAH72M8YpXL4c+MtfuLt2DFDnK4hxlf+PQpfBgf1t\n+lFvEQ39FvzwH0fw/Ysq8T/Xn4dylRQ9CYR6o9mr02BHqVLKaapIJOBDLROFqLIDQNugDYNRFOdT\noSafmy3C3QvL8equZuxp0UFrceL2eSWoVEvRPjjcMF6j0cDt9aHb6MAkDsRnE6VwSNU9U1Y6AqGr\nHTUaDZq0FkxLoXO4sFyFo53sUo+EELTqrFh14yy888BF2PbIlfj5kmmsznHT+UX46kxfQhGnRJ9h\nzVorNu5rxeziHDQluSwhmB2NWiysUIfVclNKhTAmabXjuHnmd3UxRfT8ET/ZVVXw1tXh9T0t6Pj4\nS3yaOx1vHWgbXrG7bBlQXQ3i8+Htgx043J5YSn2UvdramHSjmHmOHGwbxKJKNXDrrcAnnzCO1qef\nAt/8JrM/j8esxHz55VHn5lRuYmAAsFiAyZOHxx59FGhtZZzB3l7ggw8Avyj8D38IbNw4HK1LAdT5\nmmA43F5YXR7MKsrBwfbQAt53j3bi3sUVuG2ohUwyV5h0GGycrnT0U6aUjEo9tultGIyy8pBZ7Zj8\nFN99iytwuF2P//3yLH6+ZDoEfD4qc2Vo09tC9mvX21CskEAkSP/tVzQUmcyUlY4AkJ8thnbo70kI\nQZPWiukpinwB8a081Nvc8BEEtI2yRYKY6738TMnLRq5MxNrx4wq314enPjmJ/3fVNFw2JQ9NA+yj\nf/Gys0mLpTPCtw5TSoQwJintOG4Il3IEgDlz4Dp+Al/WdaHo1HHo5i3CJyd7sbNRy3x+wQVAVhba\nt+1Cu94WV81sVBoagBkzApv72/S4eHIu8I1vMEXsu3YBU6YA5eXDxyxfDnz++Sjdr1KllDvn69gx\n5t8e/PIvEjErGx97DPjjH5moYF4e89nMmcDcucCWLdxcPwbS/vSvra3FU089hccffxx1ETRAUsW4\nyv9HoMtoR4lCgqUz8kNCzA63F1/X9+PWOcMtHsI5MmyIZq9Ojuu9/JQqpega8QBpG7TBaHdHrNNJ\nVR9FmUiAn141DTMK5LhyGnNTV6plIZGvpUuXonmogDwTUEqFyM8WZVTkKzdbCMPQ37PqwsuQxech\nl4MuCbGyoFyFY50GVhGo1kErJuclLih84/nFEUsGYiGRZ9hf97SgMEeC2+eWYHqBHI0pinx5vD7s\nbx3EZVPywn6ulApgSlLaMRF7teiscXW5iItIztf06RD2dOM+TzdEUybh/psX4v4LK/HxyR7mcx4P\nWLYMuo1v46JJ6oSe90AYezU0BOq9TA43WrRWzCtVAvPmMX0n//CH4ZSjH5WKiX795jchw6VKCXdF\n90ePMs7XSK69llmQ8MILTCQsmB//OKWF92l1vnw+H6qrq/HEE09g1apVqK6uTt2XeYLSZXCgTCXF\nVdPysbNRB9+QPWsaBjCnRBFSg1WiTCztGI0OQ+waR2woU0rRFfQAsTg9sLm8UMmEMETot5gMna9I\nfLOqBC98e17gR7hSLUP7iMhXi86KqUnWrIoVHo+Hj398WdrV9oMR8PlQShiV+6aB1KYcAUbqQi0T\noZFF5IfRIUt8njecX4iahgG4PKld9ai3ufDe0S6sunEWeDwephVkpyzydbzLiHK1FPkRGpsrMjDy\npbU48Z039uPRfx9PjT5VJOdLKERPYTmuqPkAuOIKAMA1MwtxvNMYWDFM7roLpV/9B9+/sBLdRju3\nv7GNjYHI16F2pj+pSMBnnL5bb2UiXCOdLwD49a+BbduAfcOSFCUKCXeRuWPHgAULwn/24ovAn/7E\nRLuCuf12xpn86CNu5jAGaXW+ent7UVJSApFIBLVajaKiIvT2xv/WlyjjJv8fhS6jHWVKKcrVMqhk\nQtR1mwAAH5/owTdGNDZN9MseveaLW3V7P6VKCbqDWgy1DdpQqZYiTyaKWPeVak2t4OhHZa4M7frh\nB55Go0ELRz/UXCEYWUeSARTIGaHVL/bXpqzYPhhG7yv29F/LoJWTBvJFORLMKJBjd7MuruPjfYad\n6jVjVlFOIG1arpJCa3XB5kq++OTOJi2unBo+5QgM6XxlWM1XXY8JF01SY0G5Cve/dQj/PNyR3MBB\nZ2do6m4Ii9OD+oJK5P5nS8D5koqysHRGAT4fWoRUXzodPj4fF+laACAhW46yV1Dk60DrUMrRz223\nMZIXc+eOPlFODqP79eijwFDGooTLtGOkyBfAtF/62c9GjwuFTB3Yww8zdWFJJq1PXYvFAplMhk2b\nNmHTpk2QyWSw2WxjH0iJSJfBjjIVs4puyXQm9dhjtKN+wIIl00MfcKUJph2j0aHnVt3eT6kqNO3Y\nNmjFpFwZ1DJRxLqvVEa+RqKSCpHFY2qC/DTrrBnlfGUi/qL7XjtJeeQLYJyvnY3amH9QW3U2TObo\nb3rj+UX44nRqX0LP9jHOlx8Bn4/JubKkawECTEuhK6aFTzkCmVnzdaLbiAvKVHjwksl4/d6F+PRk\nL57b1pA8ByxC5OtIhwG26bPAczhCVjp+o6oY/6nrASEEW+sH0H39N8CrrmbqquLNdmzYAEn3CEmf\noMjX/rZBXDwpyPm65RZGwiFSKv5732P+f0j3i7PfI6uVWQgwezb7Yy+8kFko8MMfAu+/n/hcopBW\n50sul8Nms+Hee+/FPffcA4fDgYpwodUhgr1ujUbD+fbIayX7esnY9ke+NBoN5MZ27GjU4j8nezE7\nx429u3aG7H/26H4MWJzw+Hyc2svh9kJnceLMkX1Rj49nu2xI68u/3TZow+TcbHgseuw+XDtqf6+P\nwOby4vC+XWn7+1SqZfjw693QaDS44sqr0Km3o/3koYz4vmTqtsukw54jdbAKcjAtX57y6wt6T6NH\nq8ebB9pi2v9M9yD6Gk5wcv1rZhZid9MAPt9Ww/p4f00O2+vvPNkMn64j5HO51xKo+0qWvTv0Nlic\nHvScPhxxf6VUiEGLIynXj9deu0+3w6ttBQBMzs3GPaUW7K3vDDhgXNvLcuYMDvX1jfr8cLse2Qvn\nwVFQAE2QTpWh8Rj0ZitO95mx9Ww/+i5aAPtbb6FUIUa3wc76+vs3bwZ55BFcUl8f+Hz7tm2MzMSU\nKXj/ixoYLHZMH9LjCxyvVEY+/44dwPr1wK9/jZ2ffIITB/fA4mS0vhKy14kTMJeXQ7N7d3zHL16M\nQ2vWwLViBVOX1tKS8N8vLCSNeL1esmrVKqLX64lOpyNr166NuO/WrVtTOLPxy7K/7SP1fWZCCCFe\nn4/ctGEnufbPO8jJHmPY/W95ZRfpMdo5nUPTgIXc+fpeTs/px+vzkcufryF2l4cQQsh/f1BLvjjV\nS57fVk82728btb/e6iTX/ml7UuYSK6s/OUk+rO0ihBDSorWQO/6yJ63zGQ+8urOJvLyziVz5goaY\nHe60zKHXZCc3v7yTbG8YiLqf1ekmlz9fQzxeH2fX/q/3j5OPT3Rzdr6xuO213aRVZw0Z27y/lTy3\n7WxSr/vOoXay5rNTUffx+Xzk4ue+Ji6PN6lziRWP10eWvKghBpsrZNzscJPlbx0kf9x6lvh83H0X\nCCGEqFSEDIz+Ht6zcT85cbaTkG3bRn32l93NZMXbh8jtr+0mPq+XkOnTyZePPUPe/fww++svW0bI\nzTcTcuGFw2ONjYRMmkQIIeT9Y53kiY/r2J+XEEIeeYSQnBxCbr2V/PW7K0nbqabYjzWZCLn1VkI+\n+mh47OWXCXnoofjmEkxjIzO3vDxCrr+ekF272B1vNkf1W9Ia+eLz+Vi2bBnWrFmDZ599FjfddFM6\npzO2p5rhEML0kysdSjvyeTxcOS0f+dkinB+UUgimVCEZtXowViLZq0NvQ0US6r0AvxjfcHi6ddCG\nSbky5GaLoA9T82V0eNJeTF6plqFtaMXjR9sPZMxKx0wmXy5GbZcBEp43rPZTKijKkeAPt8/Fms9P\no0kbuficqTuUcdohYHGlGnU9JtbHxfMMM9jdMNrdo8oEphXI0TiQ3LTj7iYtLo9S7wUwNZQKiSAp\nqcd47NWisyIvWzTquSIXC/Dnu+bjSIcBX53hUCzXYmH6N+aFpmYNdje6jXbMmlbCiJeO4NY5xTjW\nZcR1s4rA4/OBP/8Zsw9q8M07lzIpuXDtdsJx5AgjGfHOO/CcOgX4I3DB9V5telwUXO/FhvXrgZYW\n4P77saCtDrl33zlKAT8sRiNwww2M0OvPf84I0QLRi+3ZMG0aM7fOTkaW4q67gMHB2Ob17LMB4dlI\npL3Sdv78+Vi3bh3WrVuHefPmpXs6nONwe/H6npaUXEtndUEmykK2aPjHavnFk/CbG2ZFXALPqbDd\nEB0GO8qTUO/lp0zFLBTw+gg6h1ZV5maHL7g32t1JV7cfC3/RPQD0OUDrvWIgXy7CsU4jitPcbrKq\nVIlHlk7H05+ejrhPi86GyRyvXq1Qy9ChT03969k+M2YW5owSRJ6eL4/qdCaK1eVhCtcnq8fcVylJ\nXtE9W+p6TKgqUYb9LEcixF0XlGFnk5a7C/qL7Uf8fY606zG/TAVBVvif8VKlFPcuqsA3qpiei7jp\nJjRu+id+84aGcSpWrhx2WKLxxBPAqlWASgX9okVMuyAgUO/l9REcHFnvxZa8PODuu/H5ky/C7fYA\n770XfX+9Hrj+eqbR+GefMbVaf/wj81m0Yvt4kEiABx8E7ryTWaUZjT/8gXHazpwZVviPQNqdr0wi\nGTpfX9cP4PU9rQHJh2TSZbCPao5cppJiXln4BwWQ2AqTSPbq1NtRoUqelIK/x2OvyQGVVAipKAu5\nMhF0YQruU6XxFY1KtXRYbiKnIGNkJjKZ/GwxPD6CC8+rTPdUcP15hWjWWSPew60crXQMplItDduW\naizieYaNLLb3UyAXweMlUQWME+FAqx5VJYqQl8VIJKvFUDz2qusxYk6JIuLnl0zJxb7WQe6e+RGK\n7Q916LG4UhX10JXXzMDk3OGXvVKFFJ0WN+O4XHop8Le/Rb/2zp3A6dPMCkAABcuXM4r1QCDyVd9v\nRm62iJNWcqVqGbav+C+m1sod4e/t8TANs6+4gpGM4PGA555j/ru5mWm3NH9+wnMZxbPPMsKxe/aE\n//z994HXXwf27gXeegs4//yop6POV5L5uK4bXkJS8tbWZWQ0vthQqpCg28StTk2HwZbcyJdSgi6D\nHW16W6BFT65MGDbyZUrjSkc/FSoZugxMpC7TZCYylYIhzad0yEyMRCLMQrYoK6IT0sLhSkc/JUoJ\ndFYXnB4OGw1H4Ey/GbOKRtuZx+NhekE2GpMU/drfOojLpkZe5RiMUipMWnNttpzsNqGqNLLzVaqU\nQiUV4kxfYk3SA0RyvtoNTCsfFpQoJegxDUnf/PrXjNPiiuBcE8I4Qb/9LaMODzBOz5dfMg7QkLr9\n4Q4DFrOcR8T5KSQ4MHMx0xYokmO4eTPTPHzduuFo4KRJTOrxO99hpCRywpfZJIRSCTz/PCPGOtIx\n1OuZ6//tbyGK/9GgzlcQidR8tQ5a0dAf+pDqMtjROGBFqVISth6Ja5iVjuyaNZcksLw3kr06Dfak\n1XwBGFrxaEerLsj5yg4vNWF0uKGQpqdmyI9UlAWlVDg0Z0vImyglPLnZQvAAGNoip/tSSbFCEnGJ\nfqvOiikc/00FfD6KFWJ0Gdjdm/E8w870mjGrKLwzMS0/eXVfA1YnSpWxPScUEkFSIl9s7WVzedBl\ndIzZjuvSKXnY1xJDfVAsdHaOcr60Fie0FifOK2TnZMjFAoiz+Iz0zUUXMY7CO++E31mjYeq7/JIQ\nADT19YxjtG8fk3acPh3tg9y9UF5QrsJtVSXA738PPPMMU+8WjMMBPP00oxE2spTmV78C+vu5qfeK\nxHe+A5SUMFGw4Mjm448Dd9wR2th8DKjzxQEf1/VgxduH8fgHtSHNrP9T14Mbzy9CUY4kaaH7YBiN\nL5aRLy6F7QA4PV5oLS6UKNg5gWxgtL4caB8MjnyJoLe7R4X6jQ5P2iNfAFN0v7dlEDlCxhmjREfA\n5+PHV0xFYfK+RqyIdJ94hpqkV+Zy/7KRirovi9MDrdUVscn79AI5mpMU+TI73MiJseeqUiqE0Z7+\nmq/TvWZML8iGMEKdlZ9Lp+Rib0t8QrmjCBP52nK8G1dOy49rkUeI1tevfw2sXRsQOg3hd79jIl9Z\nI55XN98MfPwxo6U1dSqzyCtGJ3osihUSXDIlD1i0CFiyhJlbMK++yqQUL7109MEyGfD220zromTB\n4zFz+PBD4OKLga+/BrZuZf73+9+zOhV1voJgm/+3u7z47aensHl/G1797kLMLlbgzf2MLpCPEPzn\nZA++WVUSMSXGNd3G0TVfY1GUI4bW6ozYFzEa4ezVoWdEXiMVgXIBI8ZnR5veGogiCbP4yBZljUrv\nZkLBPQBMUsugaRjA7LLY0iwU4KFLJ+O6a65O9zQARI58dRrsKMwRQyzg3qGuUEsDCzVihe0z7Gw/\n40xE+hGflp8dscfj7mYt/n2si9X1gjE5PFCIY7s3lRlS83Wixxix2D6YBeUq1PdbYOZizh0dIer2\nvSYH/nW0Ez+5MvpqukiUKoNWuF9zDaBQMMruwezdCzQ1AffdFzK8dOlSRjz1jTeA4mJAwmRO2GZc\nYm2G9/cAACAASURBVOL3v2eico8/zqQ5zWYm4vXss5GPufJK4OokPzOmTGHEY1euZGrhbruNcchY\npjqp85UAa7eehdPjw5v3L8b0AjkeXTod1Uc70Wmw42CbHkqJEOcV5URcicc1/r6ObBBm8ZErE6Hf\n7ORkDi06KyYlOa2mkAjB4/Fwqscc8saeG0bl3mh3Q5UBfQsrc6U40mGg9V7jlBJF+FXByVjp6KdC\nlfzIV6Riez/T8rPRrA1dbEAIwTuH2vHsF2fw1z0tONyhj+vaZoeHVeQrE2q+TnabUBWl2N6PRJiF\n+eVKHGyLzzYhjIh8rdc04u4F5SiOM7sQssiKxwNWr2YciYaG4Z2efZZJ4wnDPDsvvhjwegMrHXvN\nDpQkw/maNAk4eBA4fpxx+FavZhYKhGtXlGr4fEZ+4vRppkflzTezP0USpjVuYZv/P9FtxIrLJkM2\ntFqnWCHBfRdW4oWaBnxc14NvzmV6KaplIgxGaPrMFU6PFwa7O1CozIZ42zqEsxfTYDj5q/nKlBJ4\nCQlZYRPOyTVlwGpHgEk7egmBU9sx9s6UAJmivVeqDL8whVnpmByHulItRYeBnfPF1l6ne6M7XzkS\nIRQSARr6LbA4PbA4Pfi/L8/i4xM9+Nu9i/DkTedj9SenYIojwmN0xL4YJhNqvgghqOsxRV3pGMyl\nU/Kwp5WD1GNQzdfhDj1O9phw/0XxrwL2Zw4C3HIL8OSTwNKlQG0to5N15Ajw0EOjjtVoNIBAANx4\nIzB9OgYsTqikwqREfgEAubmMlERVFfDSS0wdWCYhEoVPgcZAeiuRxzF2lxd9ZueoWon7Flfink37\n0W9x4vFrZwIA8mQinOZq5UsEeowOFCvEcdUAlCikQzdj4itWWgatuGIM0UQuKFVK4SMI0SaKFPlS\nxPh2nUwqh74nRRlSw0RhR7FCgt4waccWnZWzlV4jqVDLWKcd2XK23zzmD/lFk3Lx438dhc9H4CUE\nl0zOxev3LUK2SIASpRRLZxTg/748i//95pyIeoIjcXl88PoIJMLY3v+Z5trpjXz1mZ3wEYLSGKM8\nl07Oxd8PtoMQErNdRmEyMVEmlQoenw/rtjXgkaXTIRHG7+yUKiXY0ThCh2zFCiZtdv31jHDqf/0X\no28ViVWrAK83rLwR5wgEzCrDJ55gnLEJQvp/lTIINvn/Rq0FU/KyIeCHPjxEAj5+c8MsHGjXB9Jd\nall49XUu6Uqg6LE0TqHVcPZq1dlw/4XJj3yVKiUQZIU+0HJlIuhG2FlrdUEtEyV9PmNRqpQgVybE\nnTfE95Z0rpIM7b14YJboO0b9kDZprbh7YeR+tIlQrJDAYHPD4fbG/GPLxl52lxfdRgemjpEKf+rm\n8/HUzZE1i3521TQs//shfHKyF9+oKonp2iYH81IUq1PCNNfmvuCejb1qu4yYW6qMec6TcmXI4vHQ\nrLPGL5niTznyePj8ZC9yJAJcO7MgvnMN4ddJHMXddwNyOeNY/ehHYY8N2KuqCgDQdaKbs2L7MZlA\njhdA045xU99vibjceFGlGj+5YrgYMpIMApd0GRxxv4H4f1gSxesjaA/S3komS2cU4ObZxSFjudnC\nEDvrrC64vT4UcSD+lygCPh+f/OTymAQlKZlHtohZom8IqjvyeH1oG7RhWpLaRWXxeShVStBpSE70\nq77fjKl52QkvjpEIs/D0LbOxYUdTzMeYnR7ksFgIo5Qmp+CeDYfa9VhYEXuUk8fjDa16TEBywq9u\nD+BMnxlLphfEH0UbomQoihtWBPbWW5m0ozw2Z7Hb6Ig5EkgJhTpfQbDJ/zf0WzCjMLYvaCpWO3YZ\nmVWG8RDxTWgEhBBGnG+IkfbqGVKcl6XAwbigXIUrp4WmN3NHRBjP9ptxXmFOwg8rrhDw+RlTwzRe\nyCR7lYy4T1oHbSjKkSSUAhoLJvUYe90XG3sdaNNjQUV0hfRYOa9QDrePQGuJbeGOye6GgkXPTiby\nld6ar8MxKMqP5KJJuTjcnkDRfVCxfZ/ZiWIOXiQlwizIxYKwHUHGYqS94pE3ojBw+iu5YcMGdHd3\nQyQSYcmSJYEQZW1tLd577z3YbDY88MADqBoKWUYaHw/UD5hx/azCmPZlCsGT+9bWZbBjfunYS6DD\nUaIYUYAZgS/P9GFvyyB+e8vssJ+36qxpXc2Xmx2adhxrJReFwoaSIbkJf8F144AFMwqS+32viLPN\nUCzsatbiZ1dN4+RcPB4P5xXKcbbfjPwYFv2YnB4oWCyEkQj58BHCKgXLJf1mJ4wOD6aPIa46kgvK\nVXj2izPw+kh8jdeDnK9+s4OTFj7AUNG9wR7XAq1gaOQrfjh1vng8HlauXIn8/OGIhM/nQ3V1NZ58\n8klYrVa8+OKLqKqqCjs+Z07sBZvJINb8v48QNA1YY458yYRZ8BECu8ubFIHNum4jaruN+O/rz4vr\n+KIcMXRWFxxuL072mLCrWYdvzy9FhTo0fbizSRey9H2kvZK57D4W8kYU3J/pM+OaBOsjuCZTapjG\nC5lkr5FyEw0DsUe/42WSWoZTvaaY94/VXjqrC+16Oy4o5ybyBQAzC3Nwts+Cy2NYcGN2uJHDIvLF\n4/ECzbW5dL5itdfhDj0WlqtGNR8fi7xsEXKzRWjSWjCTpRo9AEaxfdEi5j/NThTmcOPoMKt3HWDb\nAXGkvRKpNT7X4TztSEbkkXt7e1FSUgKRSAS1Wo2ioiL09PSEHe/t7eV6OpzQogsVGewy2JEjEcQs\n3snj8aBOUurRYHPh1x/X4Tc3zIr7LUaQxUeBXIxbXtmNFzWNONljxEcnekL2IYTgYJs+am1Y6yD3\nbVbYMFJq4kyfGefRyBeFI/x98fw0DFhYR0LYEo/QaizsadHhoknqMZXa2eCPfMWCyeFhLX6skCYn\n9RgLh9v1ca9qXVCuwtFOQ3wXfuUV4KGH4Pb6YLC7kZfNzeKhEqU0kO3wEQKbi/1iBofbC7PDk3D0\n7FwlrjuvtrYWa9asCflfW1sbJBIJ1q9fj9deew1aLbOU1WKxQCaTYdOmTdi0aRNkMhlsNlvE8XQS\nLv9vd3lx9xv70TY4PLeGfvZvMSPrkbjA6yN48pNTuP68IiydkViE53+/OQd/X34h3vr+hXhkyXRs\nH7EUuUlrhUTIh83lhd3FtFAaaa9WnQ2T0hj5YqQm3CCEwORww2Bzo1KdvvmEI5NqmMYDmWSvkZEv\nJu2YbOeLndBqrPba3aTF5TE2tY6V84pycDZGSR0Ti9ZCfpjIF7fOV6z2Yort44sSLihX4mhHnM4X\nAPD50FqcyM0WxZe6DEOpUoJtZwfw8+pjuPbPO3HPpgPhC/BHEGyvHlP88kaUONOO8+bNw7x580aN\nPzQkylZXV4ctW7bg4Ycfhlwuh81mw4oVK0AIwcaNG1FRUQGtVht2PBoajSYQ9vR/CbjcPnbs2KjP\np82/CATAhk/24ZYKPpYuXYr6AQtENi2r+RC7Gdv3H0HVt67mbL5fdRE4+Ur89KqpCZ9PW38UWgCl\nS5didokCOpMV1Z/XYNlNzHz/se0gKkQEAr4MPSYH2usOhtirpqYGDX0kEPlKxt8nlm0Bnwery4t3\nv9yFIjEJpAnSNZ+R234yZT6Zvu0nE+bTbSPoMTHO/Cdf1cBsJ4Eepsm6/lVLlsDs9OCLbTUQZ/E4\nsZfH68OepgFcJtMCc0s5m6+PEAza+LA4PTi0d1fU/U83tiJXzAMwNebzuyy+QOQrld+vXpMDBqsD\nHXUHMf1q9s/vBeUq/PHL06ipqcHVcRwPAJ/v2Aepb/Rip3j//eitR6WA4OYFc/HMrbPxwMZd+NvH\nGjx8W/T5BV//tIGgVKnmZD4TdTsqJAnU19eTN998kxBCiNfrJatWrSJ6vZ7odDqydu3aqOOR2Lp1\nazKmOiZHO/Tkpg07yY0bdhK3x0sIIeQX7x8nW8/0sTrPM5+dIluOd3E2L7vLQy5/voYMmB2cnTOY\n//3iNNm8vzWw/dh7x8hXZ/rIz949SnY1DYzaX2txkmv/tJ34fL6kzCdW7vjLHtI2aCWb97eRP249\nm9a5UCYWJruLXPWChvh8PnKwbZCsePtQSq579xv7yJleE2fnO9g2SO7ffICz8wXz4FsHyeH2wTH3\ne+o/J8nHJ7pZnXvNZ6fI+8c6451a3PznRDf5nw9PJHSOb7y6i7ToLHEf/8Xp3oTnEI1/HGonT3xc\nx+qYfx3uIP/3xZkkzWhiEM1v4bTg/rXXXkN/fz9yc3Nx31BDTj6fj2XLlmHNmjUAgOXLl0cdzzR0\nVheqSpUw2FzY1az7/+3deXxb1Zk38J9kWbtsyZYty7biJE5ixzZOQkLSQiFmgClb2qQB+ikMAVKW\nljIDZRhmaAnMDJ3O5y1NC5SUT6AfEtL37ULSSTJMgbcQYihZyeKY2M4e24ltyZY3bZZsSWf+kCVL\ntuToSle6sv18/0LHsnLug3R9dM5znoO6+QWBMhN13JYb+C602t7vQnGuIq6dRYlYOb8Abx9oxf3L\ny+D1+XH88gBevG0hDrf2RS3I2trrxOx8leBlHYJV7k932/GV2dOrKB8RlkaeDbFYBJvbizPdqc/3\nCjLplLg0MJRw/mLX4BCUUknomK3Pz1vxNZ6XHIMWGDQ43e24Yj0sm4f7smNOig7XHu9wWx+kWeLQ\nZoQjl/qxNMmSHEtKtWi4PJjwUVTddg8KU5hbdetCAzbvuwiHxwt1nBshOgaGaKdjEnhNuH/sscew\nYcMG/OAHP4BWO/ZmXbRoETZu3IiNGzdGLFfGao/FbHPj9c/OY8/pbj67HRJtqtDq9ECvkuIbtcXY\n/WUn7O4RDA6NoJRjbZM8pTShuiqxtPW5MDuFxUyXmXQ4b3Wi1zmMk102mHRKaJXSiIKs4fFqTXF/\n4hUsaHsqQ8tMxDUdTUIyLV7BchPpyPcKmqVTxJ33FS1ev/jkLO7dehjHRg/A3n+xN64diYmoKFTH\nlfdlG+KecB84XJvfKvfR4rXjeAcef/c4th+/DMYYjrYPJH2E1JJSbVJ5XxYey0xEo1VKsbxMh7+c\nskz6vPB4dQ5Sja9kTKkiq/e9cxiNHYPYcyY1g69oep3DyFdJcfOCQjR2DGL/xV6U61WctxznKbN5\nnflq7UttWQepRIyvzM7DX89b8UVbP5aXBW4+sQ7hvjg68yW0fKUUlweGYLG7BS17QaYn4+ihxOfS\nUGYiqFSbXK2vLpsb31pcjB+914Sf7zmDwSEvFhal5otJhUGDM92OKz7P7uFWagJI3eHa41nsbmy4\ndSG2H+/Aj95rwrDPn/SpHUtMSex4RLDMRGp3Fa6qMeJ/xu1yn0wH1fhKypQafO1+9Fr8/cpydKTo\nuI1QMmKYXucw9GoZFNIs3FxRiE2fXUjopju+DEKyUj3zBQAr5+nx6bkeHG7rw/KywBJeUY4iNPMV\nHq9AgVXhBzt5qmwcaO1DuV494dzNTBDtPUZiy7R4GXPk6BgYGj2vLz1fNgo1MvTEWTk+Wrwsdg++\neVUxfrvuGpy3OlA3X8/5y2O8yvUqtPe7MOz1T/o8m9sbWgaNV2Dmi9/BV7R4dds9uNqkxdv3LcWI\nz49r5+QnnU5RplPC4/WhK45i1tF02z0w8FTjK5avzMlDl809obRSuGC8GGM085WkzPvrNAm1TIKS\nXEXKBl/RWB3Dodoq37iqGF02d0LLDXlKfqvct/Y6UZbimlrXzc3H8UsDONPtwKKSQPX8wCHcE+N/\nsdeVcD4Dn3RKKRouD6AykYKGhFxBca4ch1r7UKCWpeUYLQDQq2QJpyy4R3xwDnuRp5KiQC3DG9++\nOuFizPGQSbJg0ipw3hp79osxBrvby3nmKzcs54sxhlfrz6GpK/4CtPHwhtXTUssk+PmaWjx/a2XS\nrysSibC4VIvjlwcT+v10zHxJxGLcXl2E9+KY/Rp0eyESiTgvHZMxU2rwBQA6ZTZGfAz2FEw/R1v/\n73V6QoOvqiINFpfk4qoEjvHJU/J3uLafMbT3D6X8AGuNPBs1xhxUGXNCVaXzVVI4PD64R3yheFkd\nHri9voyYgs5XSuH1MywwpGdJiKtMy2HKdJkWr6IcOY5eGkhbvhcQ+MxZnfHNfI2PV7fDgwK1LGKm\nK1WzXkELCidfevSMzopxrVQfOFw7kPO1o6EDOxouY9vhtsQ7ionxsjqHkaeURsya8xWvJaVaHLvM\n/ZxHr9+PPtcw9DwVWJ3Mqhoj3m82wz3ii/rzYLwo2T55U27wJRKJUKJVoCOOg6D50Oschl4lC/3b\nb927NKFdR7mKbNg9Xnj9k0/Hx6Pb7oFKmhX3rpRkfHupCWsXl4Qei0UiFOXIIirdn7LYsdCQGQdY\n543eoDIx2Z5MfcW5Cnj9LK2DL51SCpvbC6+P+73DYnOnfLlqvArD5JXubW4v552OwOhux6ERtJht\neHPfRbz1naX4oq0/7iXZeFhSOMO0cp4e9Wd6OFeTtzqGoVNmQ8LjaQSxzM5XYalJi9c+PTfp8zrp\nWKGkTbnBF4DA4CsFS4/j1/99fob+oRHkKZOfWs0Si5Arl2CAh6XHthQn24e7vlyPmysiDxAP7vgK\nxqvZbMPCopy09OdKglWg05WPw1Wm5TBlukyLV9FoUdV0lZkAAvcOnSI74tD4WMbHy2L3wJDi5arx\nKkbPeIzF5h5JaLkqRy7B4NAInvvvk3j25gWoNGhwS2UhdjV2JtzX8fHqtrtTFi9jrgJLSrV4v4nb\nMXp8nukYj3+5pQKfn+/FX89bJ/wsGK+zPY6MvcdOFVNz8JUrx+U05H0NDAV25PD1jSNQ6yv5wVdr\nnxNlOuHe+MZcRUStrxazPWW7p7gqyVXgp3dWQybh/wBzQnLlEmhkEixI007HIL1aBquDe9qCxe6G\nISe9M18LCtU42+OYcM5vkN3jRU4CM1/y7CxkiUW4dm4+bqk0AADWLi7BrsZOXlYUgODMV+ri9e2l\nJrw7WsIinMcbfZkPCCwdp3MArZFn49/vqMJPPjwFa4xZxeYuG6qNmfGFe6qamoOvFM18jV//7x2t\n8cWXQNJ98nlfbX3CnqFYPLrdvr6+HowxNJvtqMqQma8ssQh/M26mLpNkWg5Tpsu0eIlEIvzhoRVp\n3+WljzPva3y8LLb0z3xp5NmQZokwEGNnom2Ie4HVoP9YVY2n6uaHHi8o1MCgluHz870Jvd6EHLkU\nJ7YvNWkhFolwuG0s96uxYxBf3/R5zEPDu1Nc4yuaxaVarF1cjH99vznizMdMvOdPVVNy8FWqVaRl\n5svqHObtFHmAv3ITrb3CFjQNLjsCgW+KAEv7DZ4QoaT7DyEA5Kul6E1o5iv1JQqiKVDL0G2PPli0\nebzIkSWWyrFyXgGkksg/W2sXl+BPDR0Jvd54qR7oiEQi3HN1Kd49fhkAMOAaxo/eOwmJWIQLMUo8\nWGweFKrT//9w/Vdno9sxjIZx9ckuDQxBJcvi9W/jTDQlB1+BhPvU53wFa3zxha8dj+mo8TUZY44c\nXYOBnK/AkmNORiTbTwWZlsOU6SheAXqVLK6Zr4k5X6nLYZpMwSS1yezuEeQo+NssdHNlIU5Z7Lgc\n5ykA4SbmfHlgSOExPgBw28IinOgYxOV+Fzb8uRlfX2hA3YICnO+JnieX7mXHIIlYjBvm6SNm6erq\n6tDUZUM1zXolLaHBV0tLC5577jn89re/jWhvbGzECy+8gGeeeQYnT55MuP1KjDly9Dg8Ce3+4cLq\n8PA6utcps5Oe+XIOezHoHkl7Hkc4Y64iVOW+xWLDQtpZSEhK6VXSBHO+PILcKwrVMnTHGHzZ3InP\nfEUjk2QF6lOdjL86eywWR+rjpZBmYVWNET/Y3gCP14fvXz8Xc/PVsWe+BFh2DFpRpsPh1r6Itmaz\njZYceZDQ4GtkZARr1qyJaPP7/di+fTuef/55/PjHP8b27ds5t8dK0BwvO0sMvUoGs43fchMTc754\nXnaMM+frtMWOfRcm7jQBArNes3TKlNfqmYxeLYXdM4KPPtmLFrMdVZR4GbdMy2HKdBSvAL1aBmsc\ns+bh8XIOezHi8yM3wfyqZBSoZeiJtezoTjznK5YbyvU4MG6QEI/weHn9fvQ501NP6+4lJdDIJPiP\nVTWQiMUo16tw3hp98NUtwI7VoNqSXJy3OuHwBMpj1NfXB2a+6J6ftIQGX7W1tVCrI3f7mM1mGI1G\nSKVS6HQ6GAwGdHV1cWo3m+PfgpuOWl/hNb74kKe68m5Hi92NJ/90AlsPRi8emM4yE7GIRSIUaeTo\n8wAtZpr5IiTV4k24DxdItpcLkhIw2bKjzZ3YbsfJXFWci7ZeF2xJFN/udQ5Dq0hPPa3iXAX+7wPL\nUTC6xBkcfI2fgPD5Ge/pL1zIJFm4qjgHR9r7Q/052+NAZYbsbp/KJv0ENDY2Yvfu3RFt69atQ1lZ\n2YTnOhwOKJVKbN26FQCgVCrhcrng8/k4tcerRBsoN7Ei7t+4sok5X/wuO14p58s94sM/7fwSd1QX\nYcfxDvj8DFniyBtnq8D5XkHGXDmyikyQtbYJdmOYiiiHiRuKV4BeLYsr4T48Xma7G0U5wnw2C9Uy\n1MfM+fJCw/OxNFKJGLUluTjaPoAbFxTE/Xvh8UrHET6x5KukAJs40Op1DiNXkY3sNAwIY1kxOw+H\n2/pQN78ApdXLYGxthipNR2tNZ5NGsLa2FrW1tXG9kFqthsvlwsMPPwzGGLZs2QKTyQSr1cqpPV6l\n48pNDHv9qD/Xg78drf/CB/6XHSNzvoJHOMizs8AYw0sftqAsT4knbijHJ2d60N7vwpz8yHpebb1O\nTjeXVDHmyrHnTE/G1PciZDrLH90p7Wcs7pQDIarbBxVqYi872hMssnol18zS4Yv2vpj3R5+f4c39\nF/HYdXOixjDdxUzDiUQizNUH8r7CB19ClJkYb3lZHn78P00AgCazDdVGuufzIeHh6/jp0aKiInR2\ndsLlcsHv98Nms0EqlXJun0x9fX3om8pgx0U09DGgbh4A4PXdn+IPFxhuKNdDnp0VWssPPj+exw0N\nDXjqqadCjy2D/tAHIZHXG/94xM/Q7xKBMYZf76rHjlaGEYghEYuQDR802cDvHlkJkUiEfNEQ/mvv\nYfzjXTdGvF5rnwKz81S89CeZx55eM452+PG9r5UL8u9P1cfBtkzpT6Y/DrZlSn+Eerzvr59BKmbo\ndwUOfQ7+vKDialQbc6LG63CHH7PLZgvS37MnjqCjP7I+VPDng24vTjUeQ/9ZEa//vtjJ8IVZEfPn\nzQMMW84yrKktxqljByfEa7+ZwVBQIki86uvrIR/247zVieVleaGf+41VMKhlgr7/5heqYbW5sOv/\n78WeToa6xQsE/zxMlceTYgnYuXMne/HFF9mTTz7JNm/eHGpvaGhgTz/9NHv66afZiRMnEm6P5uOP\nP4543NQ1yO7deij0+JHfHWUrfv4J+7JjIJFLYowxtnfv3tB/uzxedt0v9jK/35/w60Wz8pV69uKf\nm9g3Nu9jDZf6md/vZ3b3CLvY62BDw97Q89452Mo27jkT8bt29wi7/pf1Ec8TyvtNXWzZz/aw/Res\nQndlSgl/j5Ero3iN+fbbB9kpsy30uMfuZst+tififhAer399v4ntPNGRzi6G+P1+du3GvVHvVbe8\n/hnrsbt5/ze9Pj+76bVPmcUW/bX/YXsDW/HzT9iRtr5QW3i8fvnJGbbtUCvv/YrX9mOX2EsfNke0\n/e5IO/vZR6cF6tGY5/77S7a7sYOt+tUe1tQ1KHR3pozx45ZwCc18rV69GqtXr57QvmjRImzcuDHp\n9ngElx0ZY2jtc6G934WvVxpwutuBmuLchF4zOGoFxvK9+E5WNeTI4fMz/L8HlocOxlbLJBMOya4s\n0uA3+y9GtB291I9qYw7k2cIfnVM8uh2bku25CX+PkSujeI0JJN0Po2L08bnRulCdg27MHT1nLzxe\nFpsHhkphlqxEIhH0ail6HB6YdGM5qoyx0YR7/pcds8QiXG3S4Uh7H26vNkb87HK/C81mG25aUIBL\nA0NYOksHYFzOl8MjaBrFXL1qwrmPJzsHcbVJJ1CPxiwvy8Nn56zo94rTeqj8dCZcFl+ScuTZEItE\nGBwawc4TnVhVY0S1MQenLXZeXp/vfK+gd/5uGV66s3rCYGu8SoMGZ7odEUc7fNHWjxWzhf8gAkBZ\nvgp18wugVfIfI0LIRHq1DL1hOx7P9gRKE8QqOC1UdfugaLW+hkZ8yM4STahSz5drynQRRUGD/utE\nJ+6sMWJegRqXYhRjtdiEy/kCEMr5YqP3fOewF/su9OKmDMjxXV6mw2fnrCjXqwVN/p9OpnQUS7UK\nnO914v1mM1bXFqOiUI1T3YkPvsLXaQNHC/H/rTHeWasceTa0imy0943dKA619mFFWR7vfUqEVpGN\nO7SJnac2k8WVC0BCKF5j8scVWj3X44BKmhWx8SgYL8bY6KHawiVrF0RJuue7wOp415Tp8EV7f0RO\nssfrw3snu7B2UTFKxh1NF/7+6na4USjgzm2tIhuK7KzRI9uAT8/2YEmpNiO+4BbnKlCqVSDXZxO6\nK9PGlB58lWgV2HaoHQsNGpRoFZhfqMYFq5OXyve9aSq2N5mFRTloGZ3Js9jd6HcNY0EhLfMRMhMF\nCq2ODWbOWR24dm5+6LSJcINuL7KzxIKWBChQT6z1lYoCq+HKdEqAAe39YwOsPae7sbBIg1KdEqYY\n5wL7/AxWx7DgOwsD9b4Cy8kfNFtwaxV/u/eTtbq2GFVaOkaOL1N+8LX/Yi++tSiwQ0UplaAoR47W\nPu5nfAHRc76EVGnQ4JQ5MPg61NqHa8ryJtT9EhLl43BHMeOG4jUm/Ighr8+Ptj4Xri/XRyw7BuMV\nKDMh7EAi2rKjPQUFVsOJRKLA7FdbH0Z8ftjcI9h+vANrFwf+RpRqFbjcPxSaGQvGq981jBy5sPW0\ngNGlR6sTvc5hnOyy4YZy4Zccg9atKMNj37xR6G5MG1N78JUrh14lxfXl+aG2ikI1L3lf1hTlWIJp\nCgAADwJJREFUfHFRadCEZr4Ot/VjeVlm5HsRQtIvfOarrd8Fg0aOcr0qYtkxSOh8LyD2siPfBVbH\nu3ZuPv7Px2dw/S8/xeo3D0AiFuFrc/UAAI08G1KJGH3jThqxCHiET7hgpfuPTllwfXk+FFLhN1eR\n1JjSg68b5hXgpTurI46DqCjUJJz3Fb7+L+SRDkEVBg3OdNvh8zN80daHFbMzI98riPJxuKOYcUPx\nGqNXSdE7ekLGuR4H5heoUJyrQMfg2ExOMF4We2bOfNncIymd+QKAv600YP/TdTjwj3X45B9uwFv3\nLo1YMQjMfgVWR4LxErK6fbjg4OvDFgturSoSujsT0OeRP1N68JWvkmLZrMjZoAqDBqctjqRf2+oQ\nftlRq8hGriIbe8/2QCWVoDhXIWh/CCHC0atksDqGwRjD2R4n5hWooZZJIJdkRZ/JETDZHghUue9O\nc8J9UHaWOGaZoFKtApfGzRZaMqCSPADM0atwrseBrsEhWumY5qb04CuaikI1znTbI0o0xCs8vyQT\nlh2BwNLjO4fasDzDZr0AysdJBMWMG4rXGIU0C9lZItg9XpzrcWDeaL2lkrCj1oLxOt/jmHA0WboV\nqGXodQ5H3ItTnXAfD5NuLOk+GK9L/UMZ8eVWJZWgQC3DLZUGSMSZ9+eZPo/8ybz/u0nSKqVQyyRR\n8yDiNeAahnvEFzpxXkiVBg1OWez0LYgQgvzR2a+zYYOv4lx5RNI9YwzNZhuqinKE6iaAwOyTRi5B\nn3OsPEaz2Rbqt1BKtcoJOx6PXerHklKtQD2K9M1aY2gTGZm+pt3gCxhdeuzmvvQYXM9uMtuxsCgn\n7gNsU2mhQQOxKHBobKah9X/uKGbcULwi6dVSXOx1wuH2ojg3kFBfkjs281VfXz9aJ0okeM4XEFlu\nwjXsxclOG64R+ItkqVYRKrRaX1+PPucwzHYPKgyZUbn9u1+dEzqxINPQ55E/03PwleSOx8C3xsyo\np7WoRIsn6+anfIcQISTz6VVSHGztQ3mBKvTlsESriJj5au6yocqo4f1otESEJ90faR9AlVEjaO0x\nIHLZEQgc27akVJuRy3xk+kroU9DS0oJt27ahqqoK999/f6h906ZN6OzshFQqxcqVK0Prw42Njdix\nYwdcLhcefPBB1NTUTNqerAqDBn9q6OD8e8H+NnfZsKrGOPmT00QhzcK9y0xCdyMqWv/njmLGDcUr\nkl4lwydnu3HtnLHyOiW5crzfFCi0WldXh199ek7wJcegAvVYuYn9F3sj+i0UrSIbPj/D4NAI6urq\n8J9/OYVlszJjyTHT0eeRPwkNvkZGRrBmzRqcPn06ol0kEuGHP/wh9Hp9qM3v92P79u3YsGEDnE4n\nXnnlFdTU1ERtr66u5uXbWkWhBqctdjDGOL9eMF/in2+puPKTCSEkjfRqKcw2T0Te1ISZL7Mdf3fN\nLCG6N0GBJrDsyBjD/gu9+MW3aoXuEkQiEUy6QN5XriIbR9oHQkVYCUmXhOZZa2troVZHXx9n43YZ\nms1mGI1GSKVS6HQ6GAwGdHV1RW03m81RX5Mrg0YGPwPnpPtMy5fIdLT+zx3FjBuKV6TgebPzwwZf\nhRoZ+l3DGPb68cnevThlsWdM2kRw2bG1zwUfYyjPkFymUq0Clwdc2PWXvRh0jwi+CWCqoM8jfyad\n+WpsbMTu3bsj2tatW4eysrKoz5fL5Xj11VdhMpmwdu1a6PV6OBwOKJVKbN26FQCgVCrhcrng8/mi\ntvNBJBLhW4tL8Nb+VvzbHVWcfrcpg/IlCCEknF4dKH8TPoiRiMUwaOTosrlhdQMamQS6DDiMGRir\ncr//Qi+um5OfMffVQNL9EAZtwDKTNiM2V5GZZdLBV21tLWpr458mXr9+PQDg5MmT2LlzJx555BGo\n1Wq4XC48/PDDYIxhy5YtMJlMsFqtUdsnU19fH1pzDo7AYz2e42nD9rMMpywmVBo0V3x+8HGz2Ybq\nopy4nz/THwdlSn/oMT2ezo/nLV6Bsjwljh7cF/FzhX8IH/z1EMrmL0TVcE/G9Le0ehm6HcP487Fz\nuLZQBKBS0P4EHzst7fjSzmA0FmEp3e/pcYr/PkYjYuPXCePU1NSEY8eORSTcB509exYHDhzAunXr\n4Pf78cILL+CZZ56B3+/Hb37zGzz77LMx22PZs2cPbrrpJk59/FNDBz4+3Y1f37M47m9c3/vDMaxb\nUZYRiaGEEDKe1+ePOFINAP7zL6cwr0CN9n4XClQyrFsRfXUi3QaHRnDn5n0QQ4T3H79O8J2OQUcv\n9ePXn11At8ONX921GLMFLkhLpqfJxi0JfRJ27dqFhoYGDAwMYGhoCI8++igAYPPmzeju7kZeXh7u\nu+8+AIBYLMbdd9+Nl156CQDwwAMPTNrOp2/WGvGHo5ew/2Ivrpurv+LzA/kS4ozZKZTp6uvrQyN9\nEh+KGTcUr4nGD7wABM54HBjCgdOX8S93LBGgV9HlyCVgDKgqFr7ERDiTVolTFjvkYj/K8pRCd2fK\noM8jfxL6NKxevRqrV6+e0P7YY49Fff6iRYuwcePGuNv5IhGL8fcr5+HV+vNYMTvvinVcetyBbcha\nBdXUIoRMHSVaBU50DKLLFTgVI1OIRCIUqmUZt5KgV0shEgHlGmRMHhqZWSYfjUwD15fnI0+Zjd2N\nXVd8rspUiSojzXrFi74BcUcx44biFZ+SXDm+aO9DsVYJtSxzZpgA4M4aI26uKBS6GxHEIhFKtQrc\nubxS6K5MKfR55M+0H3yJRCL88Mb5eHPfRdjcI5M+N5hsTwghU0mJVgH3iD8jUybWf3U2jBlwaPV4\nT9XNw98syKxBIZk5pv3gCwhUvK+br8db+y9O+ryDZzpo5ouDeHZ0kEgUM24oXvHJkWdDI5Mg22ER\nuitTxlfm5OPIgc+F7saUQp9H/syIwRcAfO9rc/FhswUXrM6oP3eP+GAeAioLMydfghBC4lVVpMEc\nun0RMiUkXGoi3RIpNTHe745cwoGLvXjtrkXwM+BMtx1H2vvxRXs/GjsGce3cfPx0FT/nSxJCCCFk\n5uK91MRUdc+SEuw80YHH323AmW478lUyXDNLhzW1JXjpjmrk0i5HQgghhKTYjFl2BAL1cX66qgar\na43440Mr8O76FfinmxfgxgUFyFVk03o2RxQv7ihm3FC8uKF4cUPx4obixZ8ZNfMFAPML1ZhfSIeo\nEkIIIUQYMyrnixBCCCEkHSYbt8yoZUdCCCGEEKHR4CsMrWdzQ/HijmLGDcWLG4oXNxQvbihe/Eko\n5+vtt9/GpUuXoFKp8N3vfhc6nQ4A0NjYiB07dsDlcuHBBx9ETU1NQu2EEEIIIdNVQoOv9evXAwAO\nHz6Mjz76CPfccw/8fj+2b9+ODRs2wOl04pVXXkFNTQ2n9urqakEPOaVzq7iheHFHMeOG4sUNxYsb\nihc3FC/+JLXbUa1Ww+v1AgDMZjOMRiOkUimkUikMBgO6urrAGIu7PfgahBBCCCHT1aSDr8bGRuze\nvTuibd26dSgrKwMA7Nu3D7fffjsAwOFwQKlUYuvWrQAApVIJl8sFn8/HqV1I9fX1NLLngOLFHcWM\nG4oXNxQvbihe3FC8+JNwqYkjR47AYrHgjjvuAAB0dnZi165dePjhh8EYw5YtW7B+/XpYrVZO7VKp\nNOq/d/ToUQwMDCR+pYQQQgghaaLVarF06dKoP0to2fH8+fNobm7GunXrQm1FRUXo7OyEy+WC3++H\nzWaDVCrl3B5LrAsghBBCCJlKEpr5euKJJ5Cfnw+xWAyTyRRKwD9x4gS2bdsGAHjggQdQW1ubUDsh\nhBBCyHQ1ZSrcE0IIIYRMB1RklRBCCCEkjWjwRQghhBCSRknV+ZoKolXj51pxv6WlBdu2bUNVVRXu\nv/9+IS8n5fiI17vvvouWlhbk5OTgoYceglarFfKSUo6PmG3atAmdnZ2QSqVYuXLltN7OnWy8XC4X\nXn755dDrXbhwAe+8845Ql5NyfLy/Dhw4gI8++ggejwdPPPHEtK6nyCVese7tdM/nFq+Zds/nBZsh\nDh06xP74xz8yv9/Pnn/+eebxeFhfXx974YUXGGOM+Xy+qO2MMXbixAl26NAhtm3bNqG6n3aJxMvv\n90e8xvHjx9nvf/97IboviGTeY5s2bWI9PT1CdV0QfLzHWltb2RtvvCFE99Mu0feX1+sNxc5isbBN\nmzYJeRlpc6V4MRb73k73/MT+Fs60e34yZsyyY7Aaf1dXV6iyvk6nC1XcD6/QH94OALW1tVCr1QJf\nQXolEi+z2Rz6fa/Xi6amJhQWFgp4FemVzHsMANgM2/uS7HsMAD744APcdtttAl1BeiX6/mKMwe/3\nw+PxQK1WY3BwUOhLSYsrxQuIfW+nez73v4Uz8Z6fjGm/7BgUrMbPtRL/TJVsvJ577jlIpVKsXbtW\ngN4LI5GYDQ0NAQDkcjleffVVmEwmrF27Fnq9XqCrSJ9k32N2ux29vb2hEzemu0TjJZFIcNddd+G1\n116DWq2G2WyGw+GY9oOLK8WLREo2XjPxnp+MGTHzdeTIERQXF6OkpARqtRoulwv33nsvvvOd78Dt\ndsNkMsVsn4n4iNfLL7+MNWvW4I033hDwStIn0ZiVlpYCCBxW/5Of/ATXXXcddu7cKfDVpB4f77GP\nP/4YN910k4BXkT7JxmvRokV49tln8f3vfx8ul2vaD7ziiRcZw0e8Zto9P1nTfvAVrMYfPAYpvLK+\n0+mMWnE/vD1opiwJ8RUvACgsLITT6RTiMtKKz5jJZDLIZDIhLiNt+IiXz+fDsWPHsHz5ciEvJS34\nfH8dPXoUS5YsEeIy0ibeeAXFurfTPZ/738KZcs/nw7QvshpejX/WrFl46KGHOFfc37VrFxoaGjAw\nMICqqio8+uijwlxMGvARr9dffx39/f3QarW4++67UVRUJMzFpAkfMdu8eTO6u7uRl5eH++67b1rv\nFuIjXgcPHoTZbMbq1auFuYg04iNeb775Jjo7O6FWq/H4449DqVQKczFpwCVese7tdM/nFq+Zds/n\nw7QffBFCCCGEZJJpv+xICCGEEJJJaPBFCCGEEJJGNPgihBBCCEkjGnwRQgghhKQRDb4IIYQQQtKI\nBl+EEEIIIWlEgy9CCCGEkDSiwRchhBBCSBr9L7yiAC07VlbpAAAAAElFTkSuQmCC\n",
"text": [
"<matplotlib.figure.Figure at 0x10c0ff550>"
]
}
],
"prompt_number": 137
}
],
"metadata": {}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment