Skip to content

Instantly share code, notes, and snippets.

@jvrsgsty
Last active December 29, 2015 21:09
Show Gist options
  • Save jvrsgsty/7727977 to your computer and use it in GitHub Desktop.
Save jvrsgsty/7727977 to your computer and use it in GitHub Desktop.
Lista 5 de Estadística Aplicada II
Display the source blob
Display the rendered blob
Raw
{
"metadata": {
"name": ""
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Lista 5 : Control 1"
]
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Transformaciones"
]
},
{
"cell_type": "heading",
"level": 3,
"metadata": {},
"source": [
"Javier Sagastuy 122627"
]
},
{
"cell_type": "heading",
"level": 4,
"metadata": {},
"source": [
"ITAM Oto\u00f1o 2013"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Comenzamos por especificar las opciones de graficaci\u00f3n de iPython para poder desplegar las gr\u00e1ficas de manera adecuada."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# -*- coding: utf-8 -*-\n",
"%pylab inline \n",
"%pylab --no-import-all"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Populating the interactive namespace from numpy and matplotlib\n",
"Using matplotlib backend: module://IPython.kernel.zmq.pylab.backend_inline\n",
"Populating the interactive namespace from numpy and matplotlib\n"
]
}
],
"prompt_number": 1
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Se importan todas las librer\u00edas necesarias para trabajar"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"import pandas as pd \n",
"import numpy as np \n",
"import math\n",
"import statsmodels.formula.api as sm\n",
"import matplotlib.pyplot as plt \n",
"import scipy.stats as stats\n",
"from statsmodels.stats.anova import anova_lm\n",
"from statsmodels.graphics import regressionplots as rp\n",
"from statsmodels.stats.outliers_influence import summary_table"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 2
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Se realiza la lectura de los datos"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"path = '/Users/jvrsgsty/Documents/ITAM/Semestre 7/Estadistica Aplicada 2/\\\n",
"ML-V2013/R/data/usoEnergia.dat'\n",
"dat = pd.read_table(path, sep='\\s*')\n",
"n = len(dat)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 3
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"a) Grafique los datos y comente"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.plot(dat.x,dat.y, 'bo')\n",
"plt.legend(['Demanda[kw]'])\n",
"\n",
"plt.xlabel(u'Generaci\u00f3n de energ\u00eda [kwh]')\n",
"plt.ylabel(u'Demanda energ\u00e9tica [kw]')\n",
"plt.title('Datos')\n",
"\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEaCAYAAAD65pvjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x10a667890>"
]
}
],
"prompt_number": 4
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Podemos ver una tendencia lineal que quiz\u00e1s podr\u00eda ser bien ajustada por un modelo de regresi\u00f3n lineal simple."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"b) Ajuste un modelo de regresi\u00f3n lineal simple sobre los datos sin transformar"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"model = sm.ols(formula='y ~ x', data = dat)\n",
"fitted = model.fit()\n",
"print fitted.summary()\n",
"print \" ANOVA\"\n",
"print \"==============================================================================\"\n",
"print anova_lm(fitted)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
" OLS Regression Results \n",
"==============================================================================\n",
"Dep. Variable: y R-squared: 0.705\n",
"Model: OLS Adj. R-squared: 0.699\n",
"Method: Least Squares F-statistic: 121.7\n",
"Date: Tue, 03 Dec 2013 Prob (F-statistic): 4.11e-15\n",
"Time: 14:19:28 Log-Likelihood: -98.334\n",
"No. Observations: 53 AIC: 200.7\n",
"Df Residuals: 51 BIC: 204.6\n",
"Df Model: 1 \n",
"==============================================================================\n",
" coef std err t P>|t| [95.0% Conf. Int.]\n",
"------------------------------------------------------------------------------\n",
"Intercept -0.8313 0.442 -1.882 0.065 -1.718 0.055\n",
"x 0.0037 0.000 11.030 0.000 0.003 0.004\n",
"==============================================================================\n",
"Omnibus: 0.528 Durbin-Watson: 2.095\n",
"Prob(Omnibus): 0.768 Jarque-Bera (JB): 0.284\n",
"Skew: -0.179 Prob(JB): 0.868\n",
"Kurtosis: 3.009 Cond. No. 2.70e+03\n",
"==============================================================================\n",
"\n",
"Warnings:\n",
"[1] The condition number is large, 2.7e+03. This might indicate that there are\n",
"strong multicollinearity or other numerical problems.\n",
" ANOVA\n",
"==============================================================================\n",
" df sum_sq mean_sq F PR(>F)\n",
"x 1 302.633136 302.633136 121.658188 4.106229e-15\n",
"Residual 51 126.866018 2.487569 NaN NaN\n"
]
}
],
"prompt_number": 5
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"c) Verifique su modelo via an\u00e1lisis de residuales. Comente."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"resid = fitted.resid\n",
"respuesta = fitted.fittedvalues\n",
"\n",
"plt.plot(respuesta, resid, 'bo')\n",
"plt.xlabel('Respuesta ajustada')\n",
"plt.ylabel('Residuales')\n",
"plt.title(u\"An\u00e1lisis de residuales\")\n",
"\n",
"plt.show()\n",
"\n",
"# qqplot\n",
"stats.probplot(resid,dist = 'norm', plot=plt)\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEZCAYAAACaWyIJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREBEJHNMBEREMsdEQEQkc0wEREQy9/8BjTn8dpLpogUAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x10de3aad0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEZCAYAAAB/6SUgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xtcznf/wPHXVVEkJqdMyC7uIengfL623cr5OLOch+aY\nbTbbHNpymBsbbow5/5icZoshkhtJjiFMzCHapuVMKorq+/vDupRK0VXfq6738/G4Ho/re7zel8O7\nT+/v56BRFEVBCCGESTFTOwAhhBAFT5K/EEKYIEn+QghhgiT5CyGECZLkL4QQJkiSvxBCmCBJ/qJI\nMjMz48qVK690rYODA3v27Mny2IEDB6hdu3aGc/fu3QvA9OnT8fLyeqXPfBnBwcFUrVo13z9HFG2S\n/IXRcHBwoGTJktjY2NCoUSN8fHx4/Phxgceh0WjQaDRZHmvVqhW///57hnPTTJgwgWXLlgEQFRWF\nmZkZqamprxTDqlWrMDc3x8bGBjs7Ozw9Pdm/f/9L32fQoEH4+Pi8UgyiaJPkL4yGRqNh+/btxMXF\nsXz5clavXo2/v3+m85KTk1WI7tXkZQxlixYtiIuL49KlS2i1WoYPH27AyISpk+QvjJKzszPt27cn\nICAAeFrG+fHHH3F1ddWXXYKDg+nZsye1atVi1qxZ3Lt3L8M9QkJCcHZ2xtHRkQ0bNugTcWRkJG+/\n/Tbly5enfv36zJw5k/j4+AzXnjt3jsaNG6PValm8eDFPnjzRf2Z2JRdfX1/69+8PQOvWrQF47bXX\nKF26NCEhIZQrV46zZ8/qz7958ybW1tbcuXMny/ulxWtjY8OoUaO4cOECkZGRmc6Ljo5m0qRJODg4\n8MEHHxAeHg7A0qVLWbduHbNmzcLGxoauXbtm98ctTJAkf2FU0hJeeHg4O3fupHv37vpjy5cvZ9Wq\nVURERHD16lW6d++Op6cnwcHBnDx5kk8++STDvZYuXYqfnx8//PAD48ePJygoSH9s4sSJxMTE4Ofn\nx08//cSGDRsyxPD9998zZ84c/P39Wbp0KStXrswx9vQloAMHDgAQGxvLgwcPaN26Ne+//z5+fn76\nc9avX8+///1vypUr98L7xsbGsmDBAurVq4dWq810vE+fPsTGxhIWFkarVq3Q6XQkJiby4Ycf0rdv\nX7744gvi4uL49ddfc/wOwnRI8hdGQ1EUunXrhq2tLR999BEjR46kW7du+uNeXl44OztjaWnJli1b\naN++Pe+++y5VqlRh2rRpbN++PUONfejQoTg5OdG6dWs8PT3Zvn07AFqtlnfeeYdixYpRv359RowY\nkSExajQa3nvvPVq2bImzszPDhw/XX5tT/Fm9TzNgwADWr1+v316zZo3+N4WsHDlyhLJly1K/fn1u\n3Lihf56Q3u3btwkLC2PGjBlUqFCBwYMH4+TkxM6dO18YixAWagcgRBqNRsOvv/7K22+/neXxJk2a\n6N8fOnSIpk2b6rdr1qxJcnIyERERODk5AeDi4qI/7urqysyZMwGIj49n4sSJHDp0iIsXL5Kamoqt\nrW2Gz3r+2gkTJuT5+zVp0oQSJUoQHByMnZ0dkZGRdOnSJdvzmzZtqv8NIjtHjhzhjTfewNraWr+v\nYcOGhIaG6n9ryu7htTBt0vIXhYaFxbO2SosWLTh+/Lh++9KlS5ibm+Po6Kjfl1b7Bjh58iTNmzcH\nYOHChVy4cIGffvqJ+/fvM3fu3Ey9crK7NrfMzc2BzK3ugQMH4ufnx5o1a+jVqxfFixd/qfs+r2nT\nply5coWEhAT9vrTyT1ocr9rjSBRtkvxFodS1a1d27dqFv78/0dHRfP3113Tu3Bkzs2f/pFeuXMnZ\ns2c5cOAAGzdupFOnTgD8/ffflC1blooVKxIWFsb333+f4d6KovDLL79w8OBBzpw5w9KlS/XX5pa9\nvT0VK1bM8AMKoF+/fvj7+7N27VoGDBjwit/+mfLly9OoUSMmTJjAzZs39c9EPDw8AGjQoAFnzpwp\nVD2kRMGQ5C8KhedLFzVq1GDTpk2sWbOGNm3aUL9+febMmZPh/LQHnsOGDWPatGm0bdsWgE8++YRH\njx5RvXp1Pv30U0aOHJnh/hqNhlGjRjF27Fi6devGkCFDGDRoULaxpN+fdkyj0eDj48OQIUMoW7Ys\nx44dA6Bq1aq4ublhZmZGy5YtX/h9X1SuSX9s7dq1lCxZkkaNGhEcHMyePXsoUaIEAF26dMHMzIwq\nVarQo0ePbO8nTI9G7cVcUlJSaNiwIfb29mzbtk3NUIQoEIMHD8be3p4pU6aoHYowYao/8J03bx51\n69YlLi5O7VCEyHeRkZFs3bqViIgItUMRJk7Vss+1a9fYsWMHQ4cOle5oosjz8fGhRYsWTJkyhUqV\nKqkdjjBxqpZ9evXqxYQJE3jw4AHfffedlH2EEKKAqNby3759OxUrVsTV1VVa/UIIUdAUlYwfP16x\nt7dXHBwcFDs7O6VkyZJK//79M5yj1WoVQF7ykpe85PUSL61Wm2MOVi35pxccHKx06tQp034wivDy\nzddff612CPlKvl/hVZS/m6IU/e+Xm9xpNP38ZQi6EEIUHNW7egK0adOGNm3aqB2GEEKYDKNp+Zsi\nnU6ndgj5Sr5f4VWUvxsU/e+XG6qP8H0RjUYjPYGEEOIl5SZ3SstfCCFMkCR/IYQwQZL8hRDCBEny\nF0IIEyTJXwghcpCYnEiqUrRWRJPkL4QQLxAeE06DpQ3YcWmH2qEYlCR/IYTIQkpqCrMOzsLdz53x\nLcfTsVZHtUMyKKMY4SuEEMbkj/t/MHDLQBQUjnsdp/pr1dUOyeCk5S+EEOmsPbOWhssa0r5me/YO\n2FskEz9Iy18IIQC49+geo3aMIvx6OLv67cKtspvaIeUrafkLIUzevqv7cF7sTLkS5Tjx4Ykin/hB\nWv5CCBOWlJzEpL2TWHd2Hcs7L6d9rfZqh1RgJPkLIUzS2Ztn6evflzfKvsGpYaeoYF0hy/MCAkKY\nPz+IpCQLLC2TGTPGnY4dWxdwtIYnyV8IYVJSlVTmH53PNwe+YcY7MxjsOjjbxaQCAkL46KNdREZ+\no98XGTkRoND/AFBtSufExETatGlDUlISVlZW9O7dm08++SRjcDKlsxDCgKIfRDPo10HEP47Hr7sf\nWlvtC8/38JhEUNC0LPb7EBg4Nb/CzDOjntLZysqKffv2cerUKfbv38+KFSu4fPmyWuEIIYq4n8/9\njNtSN1pVa8WBDw7kmPgBkpKyLo4kJpobOrwCp2rZp2TJkgDEx8eTnJyMpaWlmuEIIYqgB0kPGLNz\nDAf/OsjW97fSxL5Jrq+1tEzOcr+VVYqhwlONql09U1NTcXZ2plKlSowePZqqVauqGY4QoogJ/TMU\nl8UuFDcvTviw8JdK/ABjxrij1U7MsE+rnYC3d1tDhqkKVVv+ZmZmnD59mqioKDp06ECLFi1wdXXN\ncI6vr6/+vU6nk7U3hRA5epzymMnBk1l5aiVLOi2hy5tdXuk+aQ91FyzwITHRHCurFLy92xndw97g\n4GCCg4Nf6hqjWcP3s88+o2bNmgwfPly/Tx74CiFe1oXbF+jr35dKpSqxossK7ErZqR1SgTPqB763\nb9/m/v37ANy5c4egoCC6du2qVjhCiEJOURR+CPuBFitbMNh1MNs9t5tk4s8t1co+MTExDBw4kJSU\nFOzs7Pjss8+oXLmyWuEIIQqxG/E3GLJ1CNfjrxM6OJTa5WurHZLRM5qyT1ak7COEyMnWC1sZtn0Y\nQ1yH8FWbryhuXlztkFSXm9wpI3yFEIVS/ON4xu4ay+4ru9nUaxMtq7VUO6RCRZK/EKLQOXrtKP02\n96NF1RacHn6a0palgaI7D09+kOQvhCg0klOTmX5gOgvDFvJ9++/p5dhLf6woz8OTH6TmL4QwWulb\n8imlb3GrdTBVK73Oqq6rqFK6SoZzC+s8PPnBqLt6CiHEi6S15IOCprL/QTVCa//E7eDXGVN2YqbE\nD0V7Hp78IGUfIYRRmj8/iMiYT6B3TygbCav3cedmPRYm+9C5ky7T+UV5Hp78IC1/IYRRirGOguHO\ncFcLy47BzXpA9i35ojwPT36Qlr8Qwqg8evKIz3d/zqXa22DDZrj6dobj2bXkC8s8PMZCHvgKIYxG\neEw4ff374mznTDeLfkz89FCG3jta7QTmzZOEnhMZ5CWEKBRSUlP47tB3zD48m/+2+y99nPoAUMrC\nRlry+URa/kIIVf1x/w8GbBmABg2ru62m+mvV1Q6p0JOWvxBCVS8acbt9+34mrJ/L7w67cfi7Kd/1\nnCSJvwBJ8hdC5FlWSR7IdsRtfHIcQzd/TvxrwMpQLl13ZeyBiZhrzKWsU0Ck7COEyJOsplXQaidS\nuvQ9wsMXZTq/Ya9+nK25jcTwQfC/GZBcQn/MFEfj5gcp+wghDO75Vv6tW3eJjMyY5CMjv6Fs2YEZ\nLzRPgncmcqbmVt78vQu/Bc7LdG8ZjVtwJPkLIXItq1a+ldWAbM5Oeva24lno0Rfu1qTFGS+KPSnB\nb1lcIaNxC46qyf+vv/5iwIAB3Lx5kwoVKvDhhx/Sp08fNUMSQrzA/PlBGRI/QGJitSzPdXAoRVnb\nCVypUBFafQO7Z/FG3EU+ndceeFr/f74Pv7d3u/wLXmSgavIvVqwYc+fOxcXFhdu3b9O4cWM6d+6M\njY2NmmEJIbKR9eRp7lhZjSAx8Qf9Hq12Ah/5tOe/V//Da9E3+NcZT8pWuoL3tPYZHuhKH371GNUD\n386dOzN27FjeeustQB74CqG2rOr7WT3EdXUdSsWKlfWJ3LWvBStvLMK7sTdftvwSCzOpMBek3ORO\no0n+ly9fxt3dnd9++w1ra2tAkr8Qasqqvm9nNwQow/Xrc/T70k+58CDpAd47vTn812H8evjRuEpj\nFSIXhaa3T1xcHL1792bu3Ln6xJ/G19dX/16n06HT6Qo2OCFMRG568Vy/vgJX16E4O2cu14T+GUr/\nzf3x0HoQPiwc6+LW2XySMLTg4GCCg4Nf6hrVW/5PnjyhY8eOdOjQgY8//jjDMWn5C1EwsuvFk5j4\nY6Zz27TxJTjYV7/9OOUxk4Mns/LUSpZ0WkKXN7sURMjiBYy+5a8oCkOGDKFevXqZEr8QIn+lb+mf\nPXueO3c2ZjieXS+e9N0xf7/9O339+1K5VGVODTtFpVKV8jVmYTiqLuZy8OBB/Pz82Lt3L66urri6\nuhIYGKhmSEKYhGdLJE5j/35f7typk8VZT3vxpJe2OIqiKCwKW0TLlS3xcvNim+c2SfyFjKot/5Yt\nW5KamqpmCEKYpMz99bNaArE1der8SMWKGev7Ddr8i07rO3Ez4SYHBx/kzfJvFlTYwoCM4oGvEKJg\nZe6v7w5MBDIOupo6dUCGvve//v4rLotd8HLz4qs2X1HMvFiBxCsMT5K/ECYo82LnTxN8uXLvU69e\n7UyDruIfx/NJ4CfsubqHX977hRbVWhRwxMLQJPkLYYLGjHHPYnqFQObNG5lplO3Ra0fp69+X1tVb\nc2r4KUpbli7ocEU+UL2r54tIV08h8k9AQAgLFuxOV89vmyHxJ6cm803INyw6voiFHRbybt13VYxW\nvIxCNcI3K5L8hVDH5buX6effj9KWpVnVbRWv27yudkjiJeQmd6ra1VMIYVwURWHFyRU0W9GMPk59\nCOwXKIm/iJKavxAm5EVr6t5+eBuvbV5cvXeV4IHBOFZ0VDlakZ8k+QtRhKVP9g8eXCMmpnSGSdnS\n1tQ1+1cCQ7cNpZ9TPzb03IClhaVaIYsCIjV/IYqozPP1TAKmZTyp2EOqDW4DtW+yuttqdA66Ao5S\n5Aejn9tHCGFYL56v57n/7pVPQo9+JKaac2H4aV6zeq1AYxXqkuQvRBGRuaXv+9wZ/wzs0qRAi2+h\n2RwI/C+ur5+XxG+CJPkLUciltfbDwi5z796GdEeeH8XrDq+Ngm5nQTGDpcfRll8s6+aaKOnqKUQh\nln52znv3aj93NG2+HgAF6v+B2bCVVEl4Quuo1ng0XaZfgUuYHmn5C1GIZZydM+v5esq+3h2Lrpd4\naHOD/zT4Hu8ZQwo0RmGcJPkLUYhlnJ0z88yclVss5kmHg3i6ejLj3zMoUaxEQYcojJQkfyEKsYyz\nc6aVb3x4rfxVyvS4THz1K6zr7Ye71l2N8IQRU7XmP3jwYCpVqoSTk5OaYQhRaI0Z445WOzHdntbY\nN4jhtXGHafBOFS58fF4Sv8iSqoO8Dhw4QKlSpRgwYAC//fZbpuMyyEuInKXNzvko0YwbNQ4RUzOM\n/3aYyyCXQWg0GrXDEyooFLN6RkVF0blzZ0n+QuTBtQfXGLhlIInJiazpvoY3yr6hdkhCRTKrpxAm\n4KeIn3Bb4oauuo79g/ZL4he58tIPfJOSkrC0LLhJn3x9ffXvdTodOp2uwD5bCGMWmxiL905vjlw7\nwvY+22lcpbHaIQmVBAcHExwc/FLX5Fj28fT0ZMmSJVhbW9O8eXNiYmLw9fVl8ODBeYlVT8o+Qry8\nA38cYMCWAXhoPZjtPhvr4tZqhySMiEHKPufOnaN06dJs3ryZBg0acPHiRVasWGGwIIUQufc45THj\n/zee935+jwXtF7C402JJ/OKV5Jj8S5YsycOHD1mzZg39+vXDysqKuLg4g3y4p6cnzZs35+LFi1St\nWpX/+7//M8h9hSiKzt86T7MVzTh76yynhp2i0786qR2SKMRyrPl7e3vj5uZGw4YNad68OVFRUZQp\nU8YgH75+/XqD3EeIokxRFBaFLcJ3vy/T3prGhw0+lC6cIs9euqunoiikpKRgYZH/g4Ol5i9M3fX4\n6wzZOoRbCbfw6+HHv8r9S+2QRCFgkJr/nTt3mDlzJl26dAHg/PnzrF692jARCiGyteX3LbgsdqFB\n5QYcHHwwQ+IPCAjBw2MSOp0vHh6TCAgIUTFSURjl2Hz/6quvcHR0JCoqCoBatWrx3nvvMWSIzAwo\nRH6IfxzPJ4GfsDdqL/69/WletXmG45kXbXm2Fq9MzyxyK8eW//Hjxxk5ciTm5uYAWFhY6N8LIQzr\nyLUjuC5xJUVJIXxYeIbEn9ba799/UYbEDxAZ+Q0LFuwu6HBFIZZjy9/NzY2//vpLv+3v70+rVq3y\nNSghTE1yajLTQqbxw/EfWNRhET3r9sxwPGNr3zfLeyQmSqNM5F6Oyf/jjz9m1KhRREVFUbNmTWrU\nqMGiRYsKIjYhTMKlO5fov7k/ZazKED4snNdtXtcfy3qJxucXbXnKyiqlAKIVRUWue/vcunWLlJQU\n7Ozs8jsmPentI4oyRVFYEb6C8XvG07NcH65sLMXjpGI8eHANKM7jxylcuaLh0aPFPG3t+/5zZQiw\ni/SLtmi1E2RJRqGXm9yZY8t/9erVWfYpHjBgwKtHJoSJu5VwC69tXvwR+weTa3zHnC8v/lPSSZ/Y\nJwHT/rki60Vbypb9k8aNq+HtLYlfvJwck39YWJg++d+5c4egoCDc3d0l+QvxinZe2snQbUPpX78/\nG9/dSJeOU9M9wA3iWYv+RUs0tkarDWTevCGS9MUryTH5f//99xm2o6OjDTapmxCm5OGTh4wLGsf2\nS9tZ12MdbRzaAM+vw5v+vbT2Rf556WG6ZcqUITo6Oj9iEaLIOvH3Cfpt7odbZTdODz/Na1av6Y9l\nXIc3/Xtp7Yv8k2Py79y5s/59UlIS586d4/PPP8/XoIQoKlJSU5h1cBZzj8xlXrt5eDp56o+l9eSJ\njr5FiRLD/3mwmz7hP03wJUr0RqutTJUqNtLaFwaTY2+f9AsEWFlZ4eLigpWVVX7HBUhvH1G4Rd2P\nov/m/liYWbC622qqlakGPE36Pj4/cv58MRITf/jn7BBKlFiIVluZYsXi0WgssbGpgJVVCt7ebSXh\ni5dSKNbwfRFJ/qIwUhQFvzN+jA0ayxctvmBss7GYaZ4Opn82WEvDs548z3h4+BAYOLWAIxZFTZ66\nepYqVSrbaWM1Gg0PHjzIW3RCFEF3H91lRMAIzt48y+7+u3GxcwGyGqzlm+X1MkpXFJRs5/aJj48n\nLi4uy5ehEn9ISAh16tShVq1aLFiwwCD3FEIte67swXmxM5VLVea413GiTzzAw2MS9eoNo1evdQQF\nTePevdr/nC2jdIW6cl32iY2NJTY2Vr9drVq1PH+4q6sr8+bNo3r16nh4eBAaGkr58uWfBSdlH1EI\nJCYnMmHPBH6K+Inhdp9wYPU9oqNvpRudm36wVtp7GaUr8o9BRvju378fX19fwsLCsLKy4u7du9St\nW5ezZ8/mKbi0HyStWz/9h+7u7s7Ro0fp2LFjnu4rREE6c+MMff37UuZJOWw3vsU3Zy7/8xA3fcJ/\n0WAtH6ys/qBuXRumTOktiV8UmByndJ45cyarV69Gq9Vy8+ZN/Pz89Ak7L8LCwqhdu7Z+u27duhw5\nciTP9xWiIKQqqcw5PId3fnyHtiU7EzO/Ob8dq56u986LBmt58HSw1go8PODnn4dy4sRCSfyiQOWY\n/K9fv061atWwtrYmISGBPn36sG/fvoKITQijdO3BNdquaYv/eX+ODT1GxDozrkROJ/uEn9baT9Ma\nrTaFNWuGEBg4VZK+UEWOZR9bW1vi4uLo0KED7777LlWqVKFOnTp5/uBGjRoxbtw4/XZERATt2rXL\ndJ6vr6/+vU6nQ6fT5fmzhXhVP0X8xOgdoxnTZAxftvwSCzOLdNMzZDc6VwZrifwVHBycYUxWbmT7\nwHfTpk107tyZlJQUrKysMDc3Jzg4mOjoaLp164a1tXWeA0574FutWjXatWsnD3yF0YpNjGX0ztEc\niz7G2h5rafh6wyy6bz7/EPfZwK2nCV8Ga4mCkadBXt26dePgwYO0a9cOT09PPDw8DL584/79+xk+\nfDhPnjxhzJgxjBkzJmNwkvyFCtJPu3D9+n1KO6US3TiQ0jeqU/vPLiTcv0lsbBx//237T40/fdIP\nAXbLQ1yhqjyP8I2NjWXz5s1s2LCBU6dO0a1bNzw9PWnTpo3Bg80yOEn+Ih+lJfmkJIssFlDpA+Y7\nQKcBl2Ww9R24tJFnif75EbpPk/6zGTellS/UY9DpHW7fvs0vv/zCwoULuXv3LteuXTNIkC8iyV/k\nh8xz62SxgEqFYdAjDB7Yw1YtJMz95+q0Lpy+ZDVKt00bX4KDM+8XoiDlJnfm2NsH4N69e/j7+7Nx\n40bu3r1Lr169DBKgEAUtbW6d8HC7dN0y0y+gYg6Nv4cP1sDxYbD+V0gok+4OWT3cfUZG6IrCItve\nPnFxcfqSz8mTJ+nSpQs+Pj7odLps5/wRwli9eG6df/4blIqBrn5QshysGAh3hv1zPKv59p8frPV0\nhK63d+Yea0IYo2yTf40aNfDw8GDkyJG4u7tTvHjxgoxLCIN5NpPmNzxL+s8l9NqbodMION4OQuwg\ntQPPkrt7Nu9BRuiKwirbmv/Dhw8pWbJkQceTgdT8hSF4eEwiKCibuXWKfwntPMHhIPjvgGvNSOui\nWbGiBfHxKVSubJdhjv0HD67JfPvCqMl8/sKkpZV6jh69Rmzsqn/2pnu4a78IekzE/Fo5qp51xraU\nnSR0USQYZGI3IQqjjKWeSemOtAazZGjdGk3jMJz/6sS0wd6S6IXJkZa/KJIylnrStfZtL0GPfpTQ\n3GJx++8Y0L2HilEKkT/y1PJPv3D78zfSaDRs3brVACEKYVjpSz3PtAYUcOuI5t97ePOGjm/fnUWn\nTgUzWFEIY5Rt8v/0008B2LVrF6dOnaJ3794A/PTTTzg7OxdMdEK8hGxLPdY3ofMcKPM3zS8OInTz\nYtViFMJY5Fj2cXV1JTQ0VD+RW0JCAi1btiQ8PDz/g5Oyj3gJWZZ6arWALkPh1EDe+DOV+XM7Sn1f\nFHkGeeBra2tLREQEjRs3BuDcuXOUK1fOMBEK8YqympfnypWHz04o1hDc50CthVjvbErLqhZ4z5Ue\nPEKkyTH5z5gxg6FDh+p/ipibm7N06dJ8D0yI56WfbfPZ+rjPz8sDVD4BPftCdCP44Q9a6r4jMHCq\neoELYYRyTP6NGjXi9OnTREdHoygK9vb2BRGXEEB2CT/9+rjp5uXRvAMtddD0HOycD2fflykXhMhG\njsk/KSmJLVu2EBISwsKFC7l06RIXLlygU6dOBRGfMEE5J/z0/2z/ef/aVejhAykPYcn7lNEE0tQj\nQlbMEiIbOSb/r7/+GkVR9EuEvf7667z77ruS/IXBPF+/j4kpzfXrc8g+4aefl+cJOK8G98/g4Bdw\neCwoZjT18JFSjxAvkOOUzvv27WPmzJn6id2sra3z3ANn06ZNODo6Ym5uzsmTJ/N0L1G4pXXPDAqa\nxv79voSH2/2T+CHHBdFL3IVeh6DFZ/Dj/+DQZ6CY/VPqaVtwX0KIQijHlv+bb75JbGysfvvIkSO4\nurrm6UOdnJzYvHkzw4YNy/lkUaTNnx/0T7/8NDkk/LQF0d8IQ9PNHtsbb2J/tCPmVZZiUzttXh4p\n9QiRkxyTv7e3N927d+fatWu89dZb3LhxgzVr1uTpQ2vXrp2n60XRkZT0/D/BFyR8wKrUu1h3jSSh\nWiSTnKYx0XNswQQqRBHzwuSfmprK1atX2bt3LydOnCA1NZVGjRoVVGzCBFhaPr8iVuaEX6JEb7Ta\nytjUekh0k5M0fqMBSzrtwbaEbQFHK0TR8cLkb2ZmxowZM+jVqxcNGjR4qRu3bduW69evZ9o/ffr0\nDPMGCdM2Zow7kZET05V+WmNnt4rXXx+ln1551OgRXLQ9wYyDM5jtPpv+9fvLanJC5FGOZZ9u3brx\n+eefM3DgQF5//XX9flvbF7e6du/enffoAF9fX/17nU6HTqczyH1FwUrfo8fSMplmzV7n8OG/SUqy\noHTpG7i5jUo3l/4gfc3+r9i/GPTrIJJuJnFs6DFqlK2h8jcRwvgEBwfre2TmVo5z+zg4OGTZyrp6\n9epLfVBW3nrrLb777rtsf6uQuX0Kr+y7bwKEYGGxjuTkZxOsabUTmTfPI8OD2o1nN+K905tPmn7C\n5y0+x9zMvIC/hRCFk9Gu5LV582bGjBnD7du3KVOmDK6uruzcuTNzcJL8C6WMs2tCxv76WW0/5fFP\n3/zYxFhtbFtDAAAZR0lEQVRG7xxNWHQYa3uspcHrL1dyFMLU5SZ35tjPPykpiY0bNzJq1CgALl26\nxPbt2/MUWPfu3fnrr7949OgR169fzzLxi8Lrxd03s9p+KjHRnP1R+3Fe7Ezp4qU5OeykJH4h8omM\n8BUG9+Lum1ltA+aP+fNfu/H8ZSnLuyynQ60O+RWeEAKVRviKoi377pvPti0shj/brHCO4qOqUdFR\nw+nhpyXxC1EAVBnhK4qWrHry5NR9s2nT+hw+Mokr5U7yR439DKs5inkDZ0oXTiEKSK5G+Hbr1s2g\nI3xF0ZH54S5ERk6kX78qHDniQ2KieabumwAxcTEc3ToY20d32dH9FLXK1VIjfCFMVq57+6gxwld6\n+xi/jEsnpt+f/aya/uf9GRkwkmENhjGp9SSKmRfL7zCFMCl5WsbxxIkTGX4F12g0GWbhdHNzM1CY\nojDL/HD3qcTEzH3y45Li+CjwI0L+CGFz7800q9osv8MTQmQj2+T/6aefotFoePz4MYcPH6ZatWpo\nNBr++OMPmjdvTmhoaEHGKYxU5oe7T1lZpWTYPvTXIfpv7o+uuo7wYeHYWNoURHhCiGxk29snODiY\nffv2UbVqVXbv3k1UVBRXr17lf//7nyzlKPTGjHFHq52YYV/6+fSfpDzhq31f0WNjD75r+x0ruq6Q\nxC+EEcix5l+3bl3Cw8OxtLQEng76cnNzIyIiIv+Dk5p/oRAQEMKCBbvTPdxtS8eOrbl45yL9/Pth\nW8KW/+v6f1S2qax2qEKYBINM7zB16lROnTpF3759URSFDRs2UL9+fXx8fAwabJbBSfIvlBRFYdnJ\nZUzYM4Gv23zN6MajpQunEAXIIMn/8ePHbN++ncDAQADat29Px44d9YO+8pMk/8LnZsJNhm4dyrUH\n11jbYy11KtRROyQhTE6ek39ycjIeHh7s2bPH4MHlhiT/wiXgYgBe27wY6DyQyW9Nprh5/jcQhBCZ\n5amrJ4CFhQUajYaoqCgcHBwMGZsoQh4+echnQZ+x49IONry7gdbVZf1cIYxdjiN8y5Yti5ubG2+/\n/TaVKz99YKfRaJg/f36+ByeM3/G/j9PPvx+NqzTm9PDTlLEqo3ZIQohcyDH5d+zYkY4dO2bYJw/v\nREpqCjNCZzD/2Hzmt5tP73q91Q5JCPEScjW9w40bNwCoVKlSvgeUntT8jdPVe1fpv7k/lhaWrO62\nGvvSMu5DCGOSp8VcUlJSmDNnDi1btqRJkyY0bdqUFi1aMHv2bFJTU/Mc3Lhx46hTpw5ubm58/PHH\nPHr0KM/3FPlLURRWn1pN4+WN6V67O7v775bEL0QhlW3y//rrrwkJCWHZsmX60b3Lly8nNDSUSZMm\n5fmD3d3diYiI4Pjx4yQkJLBu3bo831PknzsP7/Dez+/hGzSFWoe6sW1CHO3bfUVAQIjaoQkhXkG2\nZZ9atWqxa9cu3njjjQz7r1y5gru7O5cvXzZYED///DNbt27lxx9/zBiclH2MQlBkEIN/HUyjki04\nPceBq5dm6o9ltfC6EEJdeSr7KIpChQoVMu2vUKGCwRPysmXL6Ny5s0HvKfLOf9tuqg9rTJflvbA7\n0oY/lpbLkPgBIiO/YcGC3SpFKIR4Vdn29mnYsCGzZs1i6tSMc7LPnj0713P6t23bluvXr2faP336\ndH2ynzJlCjY2NvTq1SvLe/j6+urf63Q6dDpdrj5b5M2CTSv59NAXPLn7NqwJ5MQjW6ysBmR5blbT\nNwshCk5wcLB+nfXcyrbsc/v2bQYNGkRERAStWrVCo9EQEhKCo6Mjq1atonz58nkOeNWqVSxbtow9\ne/ZgZWWVOTgp+xS4VCWV2YdmM3GnL0+2LYEzfYG0rr2TgJdbuEUIUfAMMrdPXFwcO3bsAKBDhw7Y\n2BhmOt7AwEA+/fRTQkJCKFeuXNbBSfIvUH/G/snALQNJSU3h8UY3ju7673NnhGBltZ7ExB/0e7Ta\nCcyb105q/kIYEYMk//xSq1YtHj9+jK2tLQDNmjVj0aJFGYOT5F8gAgJCGL9uLr87BOEQ04Tveviw\ncMGeLJdndHUdSsWKlTNN3yyEMB5GnfxzQ5J//tv4awBDN39BvE0y/LIWYhqg1T5dgN3PLzrDwuzS\nyheicJDkL/QCAkKYPz+IpCQLLC2TGTPGHWvHVNov7Uri6f6wexY8Kak/38PDB2/vtlku0iKEMG55\nntVTFA5ZJfb0STogIISPPtr1rBVvnsSx0jrMXC9R8/dOnN3xfaZ7Jiaa07Fja0n2QhRRkvwLuUyJ\nHYiMfLqmblrinj8/6NnxChHQsy/37zvw1qFmFHtSkrNZ3Pf5BdiFEEVLtoO8ROGQIbH/4/mBV0lJ\nFqBJhcYLYJAOjnrDhs2kxpfOcQF2IUTRJC3/Qi4pKeu/wvQDr5RS96BvB7C6DysOw92awNPWfdpv\nBwsW+KSr7ctDXSGKOkn+hZylZXKW+9PKNv7n/TnTzI+yJxy5ty4UUp/+lT9t3bcDkNq+ECZIkn8h\nN2aMO5GREzN1yRwysjWDfx1MyB8h7BwYwB2Xxyx4OFla90IIQLp6FgkBASEZumT++4PK/HBjNm87\nvM3cdnMpVbyU2iEKIQqQ9PM3MU9SnjA1ZCpLTyxlSacldK3dVe2QhBAqkH7+JuTinYv08+9H+ZLl\nOTX8FHal7NQOSQhhxKSrZyGnKApLTyylxcoWDHQeSECfAEn8QogcScu/ELuZcJOhW4cSHRdNyKAQ\n6lSoo3ZIQohCQlr+hdT2i9txWexCvYr1ODzksCR+IcRLkZZ/IZPwOIHPgj4jMDKQje9upFX1VmqH\nJIQohKTlX4iERYfhttSNh8kPOTXslCR+IcQrk5Z/IZCcmsyM0BksOLaABe0X8J7je2qHJIQo5FRJ\n/j4+PmzduhWNRoOTkxP//e9/s13K0dRduXeF/pv7U8KiBCc+PIF9aXu1QxJCFAGqDPKKi4vTrwU8\nZcoUkpOTmTJlSubgTHiQl6IorD69mnG7xzGh5QQ+avoRZhqp0gkhcma0g7zSEn9ycjIJCQmUKVNG\njTCM1p2Hdxi2fRgX71xk74C9OFVyUjskIUQRo1pTcuLEidjZ2REaGspnn32mVhhGJygyCOfFzlQv\nU51jXsck8Qsh8kW+lX3atm3L9evXM+2fPn06nTt3BuDhw4dMnPh0IZG5c+dmDk6j4euvv9Zv63Q6\ndDpdfoSrukdPHvHl/77E/3d/VnVdxTtvvKN2SEKIQiI4OJjg4GD99uTJk41/YrfffvsNLy8vjhw5\nkumYqdT8T10/RV//vtSrWI8fOv6AbQlbtUMSQhRiucmdqpR9Ll26BDyt+a9fv54ePXqoEYbqUlJT\n+Pbgt7Rd05bxLcezoecGSfxCiAKhygPf8ePHc+HCBUqUKIFOp8PLy0uNMFT1Z+yfDNwykJTUFMK8\nwnB4zUHtkIQQJkT1ss+LFNWyz7rf1vFx4MeMbTaWcc3HYW5mnvNFQgiRS0bb1dNU3U+8z8iAkYRf\nDyewXyBuld3UDkkIYaJk1FAB2Xd1H86LnSlXohwnPjwhiV8IoSpp+eezpOQkfPb5sPa3tSzvvJz2\ntdqrHZIQQkjyz09nb56ln38/apStwalhp6hgXUHtkIQQApCyT75IVVKZd2Qeb61+C+/G3vi/5y+J\nXwhhVKTlb2B/x/3NoC2DiHscx5EhR9Daal/6HgEBIcyfH0RSkgWWlsmMGeNOx46t8yFaIYSpkuRv\nQD+f+5lRO0YxqtEoJrSagIXZy//xBgSE8NFHu4iM/Ea/LzLy6RQY8gNACGEo0s/fAOKS4hgTOIbQ\nP0Px6+5HE/smr3wvD49JBAVNy2K/D4GBU/MSphDCREg//wLyJPUJFUpWIHxYOKWKl8rTvZKSsv4r\nSUyUgWBCCMOR5G8AtiVsmdV2lkHuZWmZnOV+K6sUg9xfCCFAevsYnTFj3NFqJ2bYp9VOwNu7rUoR\nCSGKIqn5G6GAgBAWLNhNYqI5VlYpeHu3lYe9Qohcy03ulOQvhBBFjNHO5y+EEEJdkvyFEJkMGzaM\nUqVKsW/fvgz758yZg6OjIy4uLnh5eXHjxo1c3zMmJoY2bdpQvXp1hg4dSkpK1p0Y5s+fT7NmzahX\nrx7bt2/X7w8NDcXd3R0XFxfee+89/Wd/++23uLq64urqipOTExYWFty/fx+AwYMHU6lSJZycZC3s\nTBQjZuThCVGkpKamKikpKcrUqVOV999/Xzl79qxSp04d5cyZM/pz9u3bpzx69EhRFEWZPHmyMmnS\npFzff8SIEcrMmTOV+Ph4pXv37sqmTZsynXPhwgWlcePGSkJCgnL79m3FyclJf6xJkybK0aNHFUVR\nlOnTpytffvllpuu3bdumvPPOO/rtkJAQ5eTJk0q9evVyHWdRkJvcqWrLf/bs2ZiZmXH37l01wxDC\nZEVFRVGnTh0+/PBD6tevj5+fH+fPn2fdunU4OjqydetWvLy8iI6OBkCn02FlZQVAhw4d2L9/f64/\n69ixY3z44YdYW1vTr18/jh49mumcvXv38vbbb1OyZEnKlStHnTp1+P333wEoU6YMd+7cISUlhXv3\n7lG2bNlM169btw5PT0/9dqtWrbI8T6Be0/rPP/9UPDw8FAcHB+XOnTtZnqNieEKYhKtXryoajUbZ\nsmXLS1/r5eWlzJo1S1EURXnw4IHi4uKS6eXq6qqcP39eefjwoVK1alX9tefOnVNatmyZ6Z5XrlxR\nnJ2dlZiYGOXy5ctKxYoVlZUrVyqK8jRnVKxYUSldurTSpk0bJTk5OcO1CQkJiq2trXLv3r1M31Fa\n/pmpNshr7NixzJo1i65du6oVghACKFeu3Ev/P1y9ejUREREsXLgQABsbG8LDw7M9/9GjR7nquVej\nRg0++eQT+vTpA0CjRo2wtLQEoHv37ixfvhx3d3cmTZrE+PHjmTXr2eDKbdu20bJlS1577bWX+i6m\nSpXk/+uvv2Jvb0/9+vXV+HghRDp2dnYvdf7u3buZNWsWISEhFCtWDIC4uDhatWqFRqPJdP769eup\nXbs2FStW1Jdrzp07R5MmWc+BNXDgQAYOHAhA8+bNcXd3Jz4+nujoaDp37gzABx98wAcffJDhug0b\nNmQo+YgXy7fk37ZtW65fv55p/zfffMN//vMfgoKC9Pte1CLw9fXVv9fpdOh0OkOGmWcy/bIwJeHh\n4YwYMYKgoCDKlSun329jY8OpU6deeG2TJk1YunQpo0ePZu3atfrWfXqpqancu3cPW1tbfvnlF8zM\nzChfvjwADg4OHD16lMaNG7Nlyxbc3d3118XGxhISEsK6desM9E0Ll+DgYIKDg1/uonwvPj3nt99+\nUypWrKg4ODgoDg4OioWFhVK9enXlxo0bmc5VIbyXsn37fkWrnaCAon9ptROU7dv3qx2aELly9erV\nDD1qcvLvf/9bsbOz09f0u3btmutro6OjldatWytVq1ZVBg8erK/ZHz9+XBk6dKiiKIry6NEjpW7d\nukqtWrWUNm3aKJcuXdJfHxoaqnTp0kVxdnZWvLy8lOjoaP2xVatWKZ6enpk+8/3331cqV66sFC9e\nXLG3t9c/PyjqcpM7VR/hW6NGDU6cOIGtrW2mY8Y+wlemXxZCGKNCMcI3qxphYSHTLwshCivVp3S+\ncuWK2iG8Mpl+WQhRWKne8i/MZPplIURhpXrN/0WMveYPMv2yEML4yJTOQghhggrFA18hhBAFT5K/\nEEKYIEn+QghhgiT5CyGECZLkL4QQJkiSvxBCmCBJ/kIIYYIk+QshhAmS5C+EECZIkr8QQpggSf5C\nCGGCJPkLIYQJUiX5+/r6Ym9vj6urK66urgQGBqoRhhBCmCxVkr9Go2Hs2LGEh4cTHh5Ou3bt1AhD\ndS+94HIhI9+v8CrK3w2K/vfLDdXKPjJVc9H/Byjfr/Aqyt8Niv73yw3Vkv+CBQto2rQpM2fOJC4u\nTq0whBDCJOVb8m/bti1OTk6ZXlu3bmXEiBFcvXqVXbt2ERkZyZIlS/IrDCGEEFlRVHbq1CmlefPm\nWR7TarUKIC95yUte8nqJl1arzTH3WqCCmJgYKleuTHJyMuvWraNDhw5Znnf58uUCjkwIIUyDKjX/\nL774gvr169O0aVOePHnCiBEj1AhDCCFMllEv4C6EECJ/GP0IXx8fH5ydnXFxcaF///7cuXNH7ZAM\naty4cdSpUwc3Nzc+/vhjHj16pHZIBrNp0yYcHR0xNzfn5MmTaodjMCEhIdSpU4datWqxYMECtcMx\nqMGDB1OpUiWcnJzUDiVf/PXXX7z11ls4Ojqi0+lYt26d2iEZVGJiIk2aNMHFxYWmTZsyd+7c7E82\n7ONbw3vw4IH+/eTJkxUfHx8VozG8oKAgJSUlRUlJSVGGDh2qLF++XO2QDOb8+fPKhQsXFJ1Op5w4\ncULtcAzGxcVF2b9/vxIVFaW8+eabyq1bt9QOyWBCQkKUkydPKvXq1VM7lHwRExOjhIeHK4qiKLdu\n3VJq1KiRIccUBQkJCYqiKEpiYqLi6OioXLp0KcvzjL7lb2NjA0BycjIJCQlYWVmpHJFhtW3bFjMz\nM8zMzPDw8GD//v1qh2QwtWvX5l//+pfaYRhUbGwsAK1bt6Z69eq4u7tz9OhRlaMynFatWlG2bFm1\nw8g3dnZ2uLi4AFC+fHkcHR05fvy4ylEZVsmSJQGIj48nOTkZS0vLLM8z+uQPMHHiROzs7AgNDeWz\nzz5TO5x8s2zZMjp37qx2GOIFwsLCqF27tn67bt26HDlyRMWIxKu6fPkyERERNG7cWO1QDCo1NRVn\nZ2cqVarE6NGjqVq1apbnGUXyz25A2LZt2wD45ptv+PPPP2ncuDFffPGFytG+vJy+H8CUKVOwsbGh\nV69eKkb68nLz3YQwNnFxcfTu3Zu5c+dibW2tdjgGZWZmxunTp7l8+TKLFi0iPDw8y/NU6ef/vN27\nd+d4TsmSJRk8eDBeXl4FEJFh5fT9Vq1axa5du9izZ08BRWQ4ufm7K0oaNWrEuHHj9NsREREmOzFh\nYfXkyRN69uxJ//796dq1q9rh5BsHBwc6dOjA0aNHcXV1zXTcKFr+L3Lp0iXgac1//fr19OjRQ+WI\nDCswMJBvv/2WrVu3FrnnGekpRaRHcZkyZYCnPX6ioqLYvXs3TZo0UTkqkVuKojBkyBDq1avHxx9/\nrHY4Bnf79m3u378PwJ07dwgKCsr+B1zBPYN+NT179lTq1aunNGrUSBk3bpxy9+5dtUMyqJo1ayrV\nqlVTXFxcFBcXF2XEiBFqh2Qw/v7+ir29vWJlZaVUqlRJadeundohGURwcLBSu3ZtRavVKvPmzVM7\nHIN6//33lcqVKyvFixdX7O3tlZUrV6odkkEdOHBA0Wg0irOzs/7/3M6dO9UOy2DOnDmjuLq6KvXr\n11fc3d2V1atXZ3uuDPISQggTZPRlHyGEEIYnyV8IIUyQJH8hhDBBkvyFEMIESfIXQggTJMlfCCFM\nkCR/UaTdvn2b999/n5o1a1KrVi0mTZpESkqKQT9j//79HD58WL+9ZMkS/Pz8ABg0aBC//PKLQT9P\nCEOQ5C+KtEGDBlGrVi3Cw8PZtWsXZ8+eZd68eQb9jH379nHo0CH99rBhw+jXrx8AGo0GjUZj0M8T\nwhAk+YsiKy4ujoiICKZOnYqNjQ1vvPEG//nPf/D392f16tV4e3vrz+3UqZN+Ou0RI0bQqFEjmjdv\nzrJly/TnODg4MGPGDOrXr0+nTp24evUqUVFRLFmyhLlz5+Lq6kpoaCi+vr7Mnj1bf13aOMoLFy4w\nYsQImjRpwqhRo/QLE61bt45mzZrh7OyMp6dnQfzRCCHJXxRdO3bsoFWrVhn21alTh2vXrnH9+vUM\n+9O30KdPn05YWBjBwcGsWLGChIQE/TmPHj3izJkzNGvWjDVr1uDg4MDw4cMZO3Ys4eHhtGzZMlNr\nP+39uHHjmDBhAkePHsXR0ZHly5cDT2d03bNnD6dPn2bJkiX59uchRHqS/EWRllXJRVEUKlSokO01\nu3fvpmPHjri6unLlyhX27t2rPzZgwAAA3n77bX2dX1GUTBPXPb998+ZNDhw4QJcuXXB1dWXx4sUc\nPHgQgIYNG+Lp6cnPP/9c5KYXFsbLKKZ0FiI/dOjQgS+//DLDvvPnz5OUlIS1tTVJSUn6/Xfv3gWe\nloq+/PJLDhw4QJUqVejevTv37t3Tn5e2ylWxYsVITEwEsv4B8/y+1NRUypUrl+Xc6n5+fhw6dAg/\nPz++/fbbIrUymDBe0vIXRZaNjQ2Ojo74+voSFxfHlStXmDhxIiNHjqRZs2YcOXKEx48fc/bsWY4d\nOwbAvXv3KFasGHZ2dly8eDFXayxUr16dW7duZdj3fMvfzs6OGjVq8Msvv6AoCk+ePOHcuXMoikJU\nVBTNmzdnzpw5xMTEZPihJER+keQvirRVq1bx+++/4+LiQu3atbGzs+Orr76iWrVqdO7cGRcXFyZP\nnoxOpwOgWrVq9OzZk3r16jF69Ohsl9VMX9d3d3fn+PHj+ge+aceft2jRIvbt24eLiwuurq4cPnyY\nlJQU+vfvT/369XnnnXfw9fXNds1VIQxJpnQWJuPw4cN4eXmxadMm6tSpo3Y4QqhKkr8QQpggKfsI\nIYQJkuQvhBAmSJK/EEKYIEn+QghhgiT5CyGECZLkL4QQJkiSvxBCmKD/B7MdPC1D0R4vAAAAAElF\nTkSuQmCC\n",
"text": [
"<matplotlib.figure.Figure at 0x10d9d40d0>"
]
}
],
"prompt_number": 6
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Al parecer, los resultados de la regresi\u00f3n parecen ser buenos a primera vista. Las prueba de significancia de la regresi\u00f3n tine un valor-p bueno y la $R^2$ tiene un valor \"aceptable\". Llama la atenci\u00f3n que el valor-p del t\u00e9rmino independiente $\\beta_0$ no es tan bajo como el del coeficiente del regresor ($\\beta_1$). Al ver la gr\u00e1fica de residuales contra respuesta ajustada, podemos ver un ligero patr\u00f3n en el cual los residuales son menores contra menor es la respuesta ajustada. Esto sugiere que quiz\u00e1s hay un patr\u00f3n en los datos que no fue capturado por el modelo. La gr\u00e1fica cuantil-cuantil muestra que los datos se comportan de manera similar a la normal aunque hay algunos datos un poco sospechosos. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"d) Aplique la transformaci\u00f3n Box-Cox y construya un intervalo del $100(1-\\alpha)\\%$ de confianza para $\\lambda$. \u00bfQu\u00e9 valor de $\\lambda$ elegir\u00eda para la transformaci\u00f3n? Comente."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"ydot = dat.y.prod()**(1.0/n)\n",
"l = -2.0\n",
"r = pd.DataFrame(columns=['l','SCRes','log(SCRes)'])\n",
"while l <= 2.0:\n",
"\tif l == 0:\n",
"\t\tyl = ydot * np.log(dat.y)\n",
"\telse:\n",
"\t\tyl = ((dat.y**l)-1)/(l*ydot**(l-1))\n",
"\tmodel2 = sm.ols(formula='yl ~ x', data = dat)\n",
"\tfitted2 = model2.fit()\n",
"\trow = pd.DataFrame([{'l':l,'SCRes':fitted2.ssr,'log(SCRes)':math.log(fitted2.ssr)},])\n",
"\tr=r.append(row,ignore_index=True)\n",
"\tl += 0.25\n",
"# print r \n",
"i = r.SCRes.argmin()\n",
"l = r.l[i]\n",
"if l == 0:\n",
"\tyl = ydot * np.log(dat.y)\n",
"else:\n",
"\tyl = ((dat.y**l)-1)/(l*ydot**(l-1))\n",
"model2 = sm.ols(formula='yl ~ x', data = dat)\n",
"fitted2 = model2.fit()\n",
"if l == 0:\n",
"\tyl = np.log(dat.y)\n",
"else:\n",
"\tyl = dat.y**l"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 7
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Se sigui\u00f3 el mismo procedimiento utilizado en clase para minimizar la $\\lambda$ utilizada en la transformaci\u00f3n Box-Cox. Esto es, se ajustaron modelos sucesivos aplicando la transormaci\u00f3n correspondiente con valores de $\\lambda$ entre $-2$ y $2$. Para cada uno de ellos se calcul\u00f3 la $SC_{res}$ y $\\log{SC_{res}}$. Finalmente se eligi\u00f3 el valor de $\\lambda$ que minimiza estos dos valores (que se minimizan en el mismo punto al ser $\\log$ una funci\u00f3n mon\u00f3tona creciente. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"model3 = sm.ols(formula='yl ~ x', data = dat)\n",
"fitted3 = model3.fit()\n",
"alpha = 0.05\n",
"t = stats.t.ppf(1-alpha/2,df = n-2)\n",
"SCstar = fitted2.ssr*(1+t**2/(n-2))\n",
"print \"IC: SC*<= \" + str(SCstar)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"IC: SC*<= 104.610925386\n"
]
}
],
"prompt_number": 8
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finalmente, con el valor elegido de $\\lambda$ se aplica la transformaci\u00f3n de Box-Cox, y se ajusta el modelo. Por \u00faltimo se ajusta un IC del $95\\%$ de confianza para $\\lambda$ con base en la $SC_{res}$."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"e) Grafique $y^{(\\lambda )}$ vs. $x$. Comente."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.plot(dat.x, yl, 'bo')\n",
"plt.xlabel(u'Generaci\u00f3n de energ\u00eda [kwh]')\n",
"plt.ylabel(u'(Demanda energ\u00e9tica [kw])^ 1/2')\n",
"plt.title('Datos Transformados')\n",
"\n",
"plt.show()\n",
"\n",
"plt.plot(dat.x, dat.y, 'ro')\n",
"plt.xlabel(u'Generaci\u00f3n de energ\u00eda [kwh]')\n",
"plt.ylabel(u'(Demanda energ\u00e9tica [kw])^ 1/2')\n",
"plt.title('Datos sin Transformar')\n",
"\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEaCAYAAADg2nttAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
"text": [
"<matplotlib.figure.Figure at 0x10dc9c5d0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEaCAYAAAD65pvjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
"text": [
"<matplotlib.figure.Figure at 0x10d9cddd0>"
]
}
],
"prompt_number": 9
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Como podemos ver la transformaci\u00f3n no elimin\u00f3 el patr\u00f3n presente en lso datos. Claramente sigue habiando bastante variabilidad, sin embargo a mi me parece que la tendencia lineal es m\u00e1s marcada en los datos transformados. \"A ojo\" parecer\u00eda como que el modelo ajustado a los datos transformados deber\u00eda presentar una $SC_{res}$ menor que el modelo original. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"f) Ajuste el modelo correspondiente y val\u00eddelo. Comente."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# No coincide con el reportado en las l\u00e1minas\n",
"print fitted3.summary()\n",
"print \" ANOVA\"\n",
"print \"==============================================================================\"\n",
"print anova_lm(fitted3)\n",
"\n",
"resid = fitted3.resid\n",
"respuesta = fitted3.fittedvalues\n",
"\n",
"plt.plot(respuesta, resid, 'bo')\n",
"plt.xlabel('Respuesta ajustada')\n",
"plt.ylabel('Residuales')\n",
"plt.title(u\"An\u00e1lisis de residuales\")\n",
"\n",
"plt.show()\n",
"\n",
"# qqplot\n",
"stats.probplot(resid,dist = 'norm', plot=plt)\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
" OLS Regression Results \n",
"==============================================================================\n",
"Dep. Variable: yl R-squared: 0.648\n",
"Model: OLS Adj. R-squared: 0.642\n",
"Method: Least Squares F-statistic: 94.08\n",
"Date: Tue, 03 Dec 2013 Prob (F-statistic): 3.61e-13\n",
"Time: 14:19:45 Log-Likelihood: -33.492\n",
"No. Observations: 53 AIC: 70.98\n",
"Df Residuals: 51 BIC: 74.93\n",
"Df Model: 1 \n",
"==============================================================================\n",
" coef std err t P>|t| [95.0% Conf. Int.]\n",
"------------------------------------------------------------------------------\n",
"Intercept 0.5822 0.130 4.481 0.000 0.321 0.843\n",
"x 0.0010 9.82e-05 9.699 0.000 0.001 0.001\n",
"==============================================================================\n",
"Omnibus: 1.309 Durbin-Watson: 2.304\n",
"Prob(Omnibus): 0.520 Jarque-Bera (JB): 0.730\n",
"Skew: -0.268 Prob(JB): 0.694\n",
"Kurtosis: 3.208 Cond. No. 2.70e+03\n",
"==============================================================================\n",
"\n",
"Warnings:\n",
"[1] The condition number is large, 2.7e+03. This might indicate that there are\n",
"strong multicollinearity or other numerical problems.\n",
" ANOVA\n",
"==============================================================================\n",
" df sum_sq mean_sq F PR(>F)\n",
"x 1 20.258502 20.258502 94.078081 3.613584e-13\n",
"Residual 51 10.982193 0.215337 NaN NaN\n"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEZCAYAAACAZ8KHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
"<matplotlib.figure.Figure at 0x10dc88250>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEZCAYAAABiu9n+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCJ0kSQghhNBJkoQQQgid/h9EVBi6n6eH3AAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x10de3a750>"
]
}
],
"prompt_number": 10
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Como se puede ver, las conclusiones que sacamos anteriormente \"a ojo\" son acertadas. El modelo ajustado sobre los datos transformados presenta una mucho menor $SC_{res}$. Los residuales graficados contra la respuesta ajustada parecen tener un comportamiento aleatorio y la gr\u00e1fica cuantil-cuantil de residuales corresponde con el supuesto de distribuci\u00f3n normal de los errores. Comparando con el modelo original, vemos que en este caso se presenta un valor-p de significancia de la regresi\u00f3n mayor y una $R^2$ menor. Sin embargo, los valores-p de los estad\u00edsticos $t$ de los coeficientes con menores a los que se presentaron con el modelo original. Asimismo, la disminuci\u00f3n de la $SC_{res}$ con respecto al modelo original sugieren un mejor ajuste. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"g) Construya un intervalo del $90\\%$ de confianza para la demanda media esperada si la generaci\u00f3n de energ\u00eda es de 500 kwh."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Comenzamos por calcular algunos valores que necesitamos para contruir el IC."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"xbar = dat.x.mean()\n",
"Sxx = sum((dat.x-xbar)**2)\n",
"s = fitted3.ssr/(n-2)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 11
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Ahora s\u00ed hacemos el intervalo de confianza con los valores calculados anteriormente"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"yhat = fitted3.predict({'x':500})\n",
"t = stats.t.ppf(0.95,df=n-2)\n",
"li = yhat - t*s*(1/n + (500-xbar)**2/Sxx)**0.5\n",
"ld = yhat + t*s*(1/n + (500-xbar)**2/Sxx)**0.5\n",
"\n",
"print \"IC: [\" + str(li) + \" , \" + str(ld) + \"]\""
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"IC: [[ 1.00882213] , [ 1.10848866]]\n"
]
}
],
"prompt_number": 12
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"h) Construya un intervalo de predicci\u00f3n de 95% para la demanda esperada si la generaci\u00f3n de energ\u00eda es de 1000 kwh. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"yhat = fitted3.predict({'x':1000})\n",
"t = stats.t.ppf(0.975,df=n-2)\n",
"li = yhat - t*s*(1 + 1/n + (1000-xbar)**2/Sxx)**0.5\n",
"ld = yhat + t*s*(1 + 1/n + (1000-xbar)**2/Sxx)**0.5\n",
"\n",
"print \"IC: [\" + str(li) + \" , \" + str(ld) + \"]\""
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"IC: [[ 1.10255233] , [ 1.96761742]]\n"
]
}
],
"prompt_number": 13
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Note que los intervalos de confianza anteriormente ajustados son para la respuesta ajustada por la transformaci\u00f3n de Box-Cox: $\\sqrt{y}$. Basta con elevar los l\u00edmites de los intervalos al cuadrado para obtener el IC para la respuesta. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finalmente visualizamos los datos transformados, el modelo ajustado y las bandas de confianza y predicci\u00f3n."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Visualizaci\u00f3n de los ICs\n",
"t, ci_data, ss2 = summary_table(fitted3, alpha=0.1)\n",
"predict_mean_ci_low, predict_mean_ci_upp = ci_data[:,4:6].T\n",
"\n",
"t, ci_data, ss2 = summary_table(fitted3, alpha=0.05)\n",
"predict_ci_low, predict_ci_upp = ci_data[:,6:8].T\n",
"\n",
"plt.xlabel(u'Generaci\u00f3n de energ\u00eda [kwh]')\n",
"plt.ylabel(u'(Demanda energ\u00e9tica [kw])^ 1/2')\n",
"plt.title(u'Modelo con transofrmaci\u00f3n')\n",
"plt.plot(dat.x, yl, 'bo')\n",
"plt.plot(dat.x, respuesta, 'g-', lw=2)\n",
"plt.plot(dat.x, predict_ci_low, 'c:', lw=2)\n",
"plt.plot(dat.x, predict_ci_upp, 'c:', lw=2)\n",
"plt.plot(dat.x, predict_mean_ci_low, 'r:', lw=2)\n",
"plt.plot(dat.x, predict_mean_ci_upp, 'r:', lw=2)\n",
"\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEaCAYAAADkL6tQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMVGcJBhjjIniJMEYY0zU/wMUfPOussg0DgAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x10de4d390>"
]
}
],
"prompt_number": 14
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Falta verificar un problema de graficaci\u00f3n que hace que las lineas de los ICs se vean un poco sucias.\n",
"\n",
"Por \u00faltimo, al ser tan grande la discrepancia entre el modelo ajustado presentado aqu\u00ed y el modelo ajustado presentado en la l\u00e1minas de RLS, me tom\u00e9 la libertad de graficar los datos transformados, el modelo aqu\u00ed ajustado y el modelo ajsutado en las l\u00e1minas para poder ver una comparaci\u00f3n visual."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.xlabel(u'Generaci\u00f3n de energ\u00eda [kwh]')\n",
"plt.ylabel(u'(Demanda energ\u00e9tica [kw])^ 1/2')\n",
"plt.title(u'Comparaci\u00f3n de modelos')\n",
"plt.plot(dat.x, yl, 'bo')\n",
"plt.plot(dat.x, respuesta, 'g-', lw=2)\n",
"respuesta2 = 5.822e-01 + 9.529e-04*dat.x\n",
"plt.plot(dat.x, respuesta2, 'r-', lw=2)\n",
"plt.legend(['Datos','Modelo ajustado',u'Modelo l\u00e1minas'], loc=0)\n",
"\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEaCAYAAADg2nttAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xtczuf/B/DXpwOFqITSUZlD6KDIIbpr1BRjZl9z3MGs\nNSvHGRPyY8th+GLfbWaYzWGGWRGVQ3elliQ55lDKIUWlKFLqvn5/tG7d7vuuu7qP9X4+Hh6P7utz\net91+7zv63OdOMYYAyGEEFIHLVUHQAghRP1RsiCEEFIvShaEEELqRcmCEEJIvShZEEIIqRclC0II\nIfWiZEEIgD179sDHx0es/NatW3BycsK9e/fkdi0tLS3cvn1bbueTh+zsbGhpaUEgENS776+//oph\nw4YpISqiTihZEImOHz8ODw8PdOrUCaampvD19UVCQoKqw1KYKVOmICoqSqTs6dOn8Pf3x6FDh2Bp\naamiyAhRD5QsiJht27ZhypQpGDRoEJKSknDr1i18/PHH2L9/v6pDq1NVVZVcz9e+fXucPn0adnZ2\ncj0vIZqIkgURUVJSgq+++gpffvkl1qxZAzs7OxgYGGDChAnYvHkzAKCyshK7d+/GoEGDMHjwYOzZ\nsweVlZUAAD6fDwsLC/z000/o1q0b7O3tcfr0acTFxWHAgAHo3bs39u7dK7xeSEgI3n//fcycOROm\npqbw9/cXeeSzevVqdO/eHR07dsSUKVMQHx8v3Pbrr7/C3d0dy5cvh7W1NVasWIHbt2/Dy8sLJiYm\ncHBwwJo1a1BaWio8pqCgABs2bEC/fv1gYmKCwMBA4blqP1q5efMmgoKCYGVlhdmzZ+PWrVvCbTwe\nD6tXr4aPjw/MzMwwb948FBcXS/2dHjt2DAMHDkSvXr1w4MABkW2VlZX4888/4eXlBScnJ2zfvh0V\nFRUSz1PzfkNCQmBubg5XV1dcvnwZBw8eRL9+/eDq6ipSO3r+/Dl+/PFH9OvXD97e3jhy5IhwG2MM\ne/fuhb29PZycnBAXFydyrbKyMmzfvh0DBw6Eu7s7Dhw4AGmTPdT1uwoLC4OnpycMDQ1ha2sr8rcn\nGoYRUktiYiLjOI5duXJF6j47duxgDg4O7Ny5c+z8+fPMycmJ7dy5kzHGWExMDNPV1WWBgYGsoKCA\nrVy5kpmamrLx48ezjIwMdvr0ada2bVtWUVHBGGNs+fLlTFdXl3333Xfs0aNHbPbs2WzQoEHCax04\ncIDl5uay58+fsw0bNjALCwvhtp07dzJdXV22ePFiVlxczMrKylhGRgY7efIkq6ioYBcvXmT9+/dn\n27ZtEx7z9ttvs2nTprFbt26x8vJydubMGeG53N3dhftZW1uzVatWsYKCAhYaGspsbGyE2zw8PJil\npSU7efIku3//PhswYAD75ZdfJP6uLl++zDp27MjCw8NZZmYm8/X1ZRzHsczMTMYYY5s2bWJeXl7s\nypUrLCMjg/F4PPbzzz9LPNfOnTtZq1at2KpVq9jjx4+Zv78/s7W1ZdOnT2cPHjxgO3fuZLa2tsL9\nly1bxjw9PdmNGzfYqVOnmI2NDYuJiWGMMXbkyBFma2vL4uPj2cWLF5mbmxvT0tJiVVVVjDHG5s6d\ny95//32WlZXF0tLSWN++fVl0dHSDflcVFRXM2tqaJSUlMcYYy8vLY1evXpX43oj6o2RBROzfv5+1\nb9++zn3efvtttnXrVuHrbdu2sbfffpsxVp0stLW1WUFBAWOMsfv37zOO41h4eLhw/zfeeIPx+XzG\nWHWysLKyEm4rLS1lenp67NGjR2LXFQgEzNLSkqWkpDDGqm9a+vr6rLy8XGqs27ZtY6NHj2aMMVZc\nXMzatGkjjK222jfA1NRUZmZmJrLd3NycpaamMsYY4/F4LDAwULgtNDSUTZw4UeL116xZw6ZNmyZ8\nferUKZFkMWTIEJaQkCDcfvjwYebr6yvxXDt37mTGxsbCG/qZM2cYx3Hs0qVLjDHGXr58ydq0acOy\ns7MZY4w5OjqyqKgo4fFLlixhQUFBjDHGAgIC2NKlS4Xbtm/fzjiOY1VVVUwgEDAbGxt29+5d4faN\nGzeyzz//XObf1YULF9jLly+ZmZkZ2717N3v27JnE90Q0Bz2GIiIsLS1RUlKCq1evSt0nMTERLi4u\nwtcuLi4ij4fMzMzQsWNHAECXLl0AAI6OjsLtXbp0QU5OjvC1g4OD8Oe2bdvCzs4OycnJAIDw8HCM\nHz8eXbt2hbGxMXJzc3Hp0iXh/o6OjmjVqpXwdWlpKWbPno0BAwagQ4cOmDt3rnD/hIQEWFtbC2OT\nJiEhAf379xcpc3V1xZkzZ4SvnZychD+bmpqKvJ/akpOTRfZ1dnYW/vzs2TP8888/8PPzg5GREYyM\njPDhhx8iMTFRamz29vbQ0qr+b1vzu+3Xrx8AQEdHB8bGxsjJyUFJSQkuXbok9e9UV1zXr1/H3bt3\n4eDgIIxr+fLlEjs4SPtdxcfHQ0dHB4cOHcLBgwdhYWGBGTNmICsrS+p7I+qNkgUR0bdvXxgZGSE8\nPFzqPkOHDkVKSorwdUpKCoYPH97oa168eFH4c2lpKTIzM+Hm5oZnz55h5syZ+OCDD3D9+nU8fvwY\n5ubmIs/OdXR0RM71v//9Dzdu3MCff/6J4uJibNy4UdgddMiQIbhz5w4KCwvrjGfo0KFITU0VKTt/\n/nyjuosOHDgQFy5cEL6ufd62bdvCzc0NUVFRKCoqQlFREYqLi1FUVNTg67zOwMAADg4OUv9OdcXV\ns2dPWFhY4Nq1a8K4njx5grS0NLHr1Pe7Gjx4MA4fPozs7Gzo6upi4cKFTX5vRDUoWRARBgYGWLNm\nDdavX4+vv/4amZmZePr0KQ4fPozZs2cDAMaOHYuffvoJ58+fx4ULF/DTTz9h3Lhxjb5mXl4eNm7c\niPz8fCxbtgzOzs4wMTFBSUkJSktLYWZmBoFAgNDQUDx48KDOcz148ABGRkbo3Lkzzp07h++//164\nzdDQECNHjsS8efOQkZGBFy9eSPwW7+zsjFatWiE0NBQFBQVYu3YtdHR0RL6JMxln9vf19UVkZCQi\nIiJw+/Zt/Pe//xXZPm3aNCxbtgypqakQCATIyclBdHS0TOeuz9ixY7Fu3TrcvHkTfD4f+/btE/6d\nfH19sW/fPiQkJODSpUvYvn278DgtLS1MnDgRX331FdLT0yEQCJCZmSnWCA7U/bt69OgRwsLC8OzZ\nM2hra0NPTw8GBgZyeW9E+ShZEDGffPIJdu/ejYSEBLi5uaFHjx7YsWMHJk2aBACYOnUq5s6di88/\n/xwBAQGYM2cOpkyZIjye4ziR873++vVt7777Lq5du4a+ffuitLQUf/zxB4DqxzuhoaGYNm0aHB0d\nUVFRAXd3d5FjXz/33LlzUVZWBmtra8yfPx+ff/65yD7bt29H3759MXr0aFhaWuLPP/+UeK7IyEjk\n5OTA2dkZ9+7dQ2RkpNT3JCmOGn379sXOnTuxYsUK+Pr64oMPPhDZd+bMmfj444+xbNkyGBsbY+TI\nkbh586bU31VDfrcLFy7EuHHjMH78eHzzzTfYsGEDPDw8AFQni5CQEMycORPTp09HQECAyLlCQkLg\n6emJgIAAGBsb47333kNeXl6DflcCgQAbN26Eubk5evXqhcePH2PFihVS4yXqjWOyfkVqgqqqKri6\nusLCwkKk+16NxYsXY//+/TAyMsKePXvQq1cvRYdE1MSKFSuQkZGB33//XdWhEELqoJSaxaZNm2Bv\nby/xW1BycjLi4+ORkpKCBQsWYMGCBcoIiagJJXxXIYTIgcKTxf3793Hs2DF88sknEm8MZ8+exYQJ\nE2BsbIxJkyYhPT1d0SERNVLXIxxCiPpQeLKYO3cu1q1bJ+zu97rk5GTY29sLX3fq1AmZmZmKDouo\nieXLl+O3335TdRiEkHooNFkcPXoUnTt3hrOzs9THDax6YKBIGX3TJIQQ9aJT/y6Nl5iYiPDwcBw7\ndgwvXrzA06dPMX36dJFvkm5ubrh27Zpweuj8/HzY2tqKnat79+5U4yCEkAaws7NDRkaGfE6mrKHi\nfD5fOO1CbWfPnmVDhw5lBQUFbM+ePczPz0/i8UoMtdGWL1+u6hBkQnHKF8UpX5oQZ02MleUv2HUL\nfcYAxgAWFzxdtYG9Rp73TYXWLF5X83hp69atAAB/f3/hrJaurq4wNjbG7t27lRkSIYQ0SvLmrzBw\n9lr0rFXmELhKZfEomtKShYeHh3BAkL+/v8i21atXY/Xq1coKhRBCGq0kPwdYsQIDa5UlDrbA4DN3\n0EFKR57moPm+MxXg8XiqDkEmFKd8UZzypc5xxn02CgadLcCrVXb3Ah9DEu+Ba8aJAlDSCG554DiO\nBnARQlQi53IizB2GipTxP/YCb/spRETEYfPmaJSX66B160oEBXnDz6/xE2vKkzzvm0ptsyCEEI3C\nGP7xsMXg+GyR4id5d8DrYoWIiDjMnh2FzMxvhNsyM5cAgNokDHlp3vUmQghppKvh2wEtLZFEkbhu\nNsAYOnSxAgBs3hwtkigAIDPzG2zZckKZoSoF1SwIIaSWyvIy3LU2Qp+H5cKym11bwzarGENa6Yns\nW14u+Rb64oW2QmNUBapZEELIv/75bi509NrAtlaiuBL+C3rkvIDOa4kCAFq3rpR4Hj29KoXFqCqU\nLAghLd6TvDsAx2Hwl68Wpzoz3AZMIEDfMTOkHhcU5A07uyUiZXZ2XyMwcKTCYlUV6g1FCGnRYj9+\nEx47T4uU3b+UAIt+Q2Q6PiIiDlu2nMCLF9rQ06tCYOBItWnclud9k5IFIaRFup8WBwtnD5GymM/e\nguePx1UUkfxRsiCEkEZiAgGSh1jB7WyOsKyKA549uo/2JuYqjEz+5HnfpDYLQkiLceWvreC0tUUS\nRdLmhdAWsGaXKOSNahaEkGbvZdkz5FoZwargpbAs3Uofb2QUQUe3tQojUyyqWRBCiIwSQz+Hbpt2\nIoni2vHf0PvO82adKOSNahaEkGbpyYMsdDAXXUgt/s3uGHbylooiUj6qWRBCSB3ipg4TSxS515Jb\nVKKQN0oWhJBm4+65UwDHYfieM8KymMC3AcZg1nuACiPTfPQYihCi8ZhAgJQBXTEg9aGwrEwHqHqU\nh3ZGXVQYmWrRYyhCCPnXpT82g9PWFkkUZ39YAv2XrEUnCnmjmgUhRCNVPC9BgbkRuha/mrTvsm07\n2F8vhLZuKxVGpj40pmbx4sULuLm5wcnJCYMGDcLGjRvF9uHz+ejQoQOcnZ3h7OyMVaua74LnhBD5\nSFjxCVq1bS+SKK6f/AP9MksoUSiIQtez0NPTQ0xMDNq0aYPy8nK4uLhgzJgx6N69u8h+Hh4eCA8P\nV2QohJBmoOjeLRhZ9UDtBU5j3+oNj+PX0EtlUbUMCm+zaNOmDQCgtLQUlZWVaN1afBAMPV4ihNQn\nduIgGFn1ECnLu5kKj+PXVBRRy6LwZCEQCODo6IguXbrgiy++gKWlpch2juOQmJgIJycnzJs3D5mZ\nmYoOiRCiQbKSjgMcB48/zwrL+PPfBRiD6RvOKoysZVFaA3d2djZ8fX2xZ88eODu/+gOXlJRAW1sb\nurq62LVrF/7++28cPXpUPFBq4CakRWECAVKdusDlcoGw7ElrQPdhPtp0MFFhZJpDY6coX7BgAbp3\n747PPvtM4nbGGExNTXH37l2xx1Ucx2H58uXC1zweDzweT5HhEkJUJG33ejhNWyBSdm7bCgz4ZJmK\nItIMfD4ffD5f+HrFihWakSwKCgqgo6MDQ0NDFBYWwtPTE1FRUTAzMxPu8/DhQ3Tu3BkcxyE8PBxb\ntmzBiRMnxAOlmgUhzV556RM8NTNGp1KBsOxCzw5wvFoALW2F9sdplpTadbakpAQxMTF49uyZSHlk\nZGS9J8/NzYWXlxccHR0xefJkLFiwAGZmZti6dSu2bt0KADh48CD69esHJycnHDx4EOvXr2/kWyGE\naLIzwdPR2sBQJFHcjDkE5+vFlCjUQJ01i7179yI4OBh9+vTB1atXsXHjRowdOxYA4OzsjAsXLigv\nUKpZENIsFWRdg4ltH5Ey/tsO4IVdVFFEzYc875t1puuff/4ZycnJMDExQUZGBiZOnIisrCzMmTNH\nLhcnhLRsceNdMPxwqkhZfuZl8Gz7qigiIk2dj6Hy8/NhYlLd66B79+7g8/k4duwY5s2bR9/yCSGN\nlhkfXj07bK1EEbtoEsAYOlGiUEt1JovOnTsjLS1N+NrAwAAREREoKCjApUuXFB4cIaR5YVVVSOtt\nBLvhY4Vl+W05lD0phEfoXhVGRupTZ7L4/fffYWpqKlJWMx4iLi5OoYERQpqX1J2h4HR04HS9WFiW\n8uu36FQqgH57YxVGRmRRZwO3i4sL3N3dMWrUKPB4POjp6SkzNhHUwE2IZip7+hgVXUzQ4cWr/7/n\n+xqj/8V8cFq0SoIiKa3rbFJSEsaNG4eYmBh4eHhg1KhR2LRpE27evCmXixNCmrf4he9Dv0NHkUSR\neeYIXC4XUqLQMA0alJeTk4PIyEhERUUhIyMDgwYNwg8//KDI+ISoZkGI5sjPvIxO3R1EymLedYHn\nwRQVRdQyqcV0H1VVVUhKSsLQoUPr31kOKFkQohniRztgWMRlkbKCrGswsemtoohaLrVY/CggIEBp\niYIQov5uxRwCOE4kUcQt+wBgjBJFM1BnzeLx48cSyxljcHBwQE5OjsICex3VLAhRT4KqSlzraYy+\nmSXCstwOWjDOKULrtu1VGBlR2mMoLS0tWFtbS9yWk5ODiooKuQQhC0oWhKiflJ9XwNU/RKTswp71\ncJ48TzUBERFKm+7D1tYWp06dkpgwXl/EiBDScjwvzgc6d4bry1dlZ507Y2BKLpypl1OzVOdfdc6c\nOSgqKpK47csvv1RIQIQQ9RY3dzzaGHVGm1qJIispEm6pD6k7bDOm1MWPmoIeQxGiWg9vpKJLLxeR\nspj3B8NzX6KKIiL1UYuus8pGyYIQ1Tnj0xvu0ddFyh7fuwVji+4qiojIQi26zhJCmr8b0fsAjhNJ\nFPH/NxNgjBJFC0M1C0JUJCIiDps3R6O8XAetW1ciKMgbfn7DVR0WAKDqZQVuvmGE3neeC8vuGuvA\n9H4RWum3U2FkpCGU1huKEKIYERFxmD07CpmZ3wjLMjOXAIDKE8a5/y3BgC++Re1hdGl/bobTe4Eq\ni4moXr2PoSorK/HRRx8pIxZCWozNm6NFEgUAZGZ+gy1bTqgoIqC0MBeV2hwGfPGtsCxpgBlYVRUl\nClJ3sigtLcXo0aMxYMCARp38xYsXcHNzg5OTEwYNGoSNGzdK3G/x4sWwtbWFi4sLrl+/LnEfQpqT\n8nLJlfoXL7SVHEm1uC/GoJ1JV+gIXpVlp5zEoOQH1B2WAKgnWXh4eMDX1xeff/55o06up6eHmJgY\npKWlITY2Ftu3b0dGRobIPsnJyYiPj0dKSgoWLFiABQsWNOpahGiS1q0rJZbr6VUpNY7ca8nVy5v+\n76iwLGb6cIAx2Li8qdRYiHqrM1k8efKkySO127RpA6C6llJZWYnWrVuLbD979iwmTJgAY2NjTJo0\nCenp6U26HiGaICjIG3Z2S0TK7Oy+RmDgSKXFkODVHWZ93ETKih9kwXNXrNJiIJqjzgbu+Ph4vPPO\nO+A4DuPGjWvUBQQCAZydnXH16lX897//FUs+ycnJmDZtmvB1p06dkJmZCTs7u0ZdjxBNUNOIvWXL\nUrx4oQ09vSoEBr6llMbt9GO/obffB6g9Z3TC6lkY+tX3MFT41YmmqjNZmJmZ4cSJE5g8eXKjk4WW\nlhYuXryI7Oxs+Pr6YujQoXB2dhZuZ4yJde3iOK5R1yJEk/j5DVdqz6eqinJk2hqid84LYdntzrqw\nvFOMoXptlBYH0Uz1dp01MDDAX3/91eQL2djYwNfXF2fPnhVJFm5ubrh27Rp8fHwAAPn5+bC1tZV4\njpCQEOHPPB4PPB6vyXER0hIk/fdLDJr7HXrUKrt0eCscxn2qspiI/PH5fPD5fIWcW6GD8goKCqCj\nowNDQ0MUFhbC09MTUVFRMDMzE+6TnJyMefPmISwsDFFRUdi7dy+OHj0qdi4alEdIw5Xk58Cgs4VI\nWeIQSwyOz6ZeTi2AUgfl3b59G7/88gtSU1Nx48YNcByHHj16oH///pgxY0adbQu5ubn44IMPUFVV\nBVNTUyxYsABmZmbYunUrAMDf3x8DBw6Eu7s7XF1dYWxsjN27d8vljRHS0sV+6gOPbdEiZffS4jDE\ncZiKIiKarM6axdixYyEQCPD++++jd+/esLW1BWMMt2/fRnp6Ovbv3w8tLS2EhYUpPlCqWRA5Uedp\nNuQh51ICzB3dRcpiPnkTnttOqigioipKm3X2wYMH6Nq1a50nkGUfeaBkQeRB0jQbdnZLsGmTj8Yn\nDCYQIGm4LQYn3BEpf5J3Bx26WKkoKqJKSksW0tbgrk1LSwuGhorvcEfJgsiDj08woqNXSShfisjI\nlSqIqH6y1ISuhv2CPuNmipQlbpiLIXM3KDNUomaU1mZhZmZWb62hsrIS9+7dk0swhCiauk2zUZ/6\nJhysLC/DXWtD9HlYIdx+s2tr2GYVY0grPaXHS5qvOpNF7969kZaWVucJnJyc5BoQIYqkLtNsyEr6\nhINLYXztEAYv3IzaHc2vHt2JPn4fKjVG0jLUmSySkpLqPYEs+xCiLoKCvJGZueS1NouvERj4lgqj\nkk5STahDqzuIjFoFRL0qO+PRDe782+ijxNhIy1JnstDTq67GBgcHw8PDA0OGDEHbtm0l7kOIJlDl\nNBuN8XpNaLmVJ0Lu8kXKci4nwr3vYCVGRVoimQbl7dixA/Hx8UhKSkK7du0wfPhwDBs2rNFTgDQG\nNXCTlqimzcLiiTn4BbNEtsUEjILnD8dUFBnRBErrDfW6vLw87N+/H9999x2KiopQWloqlyBkQcmi\n+Wnu4x3kgQkE4LRFG98rOaDsUQ4MTBTfZZ1oNqUvqzpjxgykp6ejS5cucHd3x6FDh0TmdyKkodR5\nWVF1ERvgC4+fjouUJS35EINW7YSBimIiLZdMyeLx48eorKyEoaEhjI2NYWJiAl1dXUXHRpqxunr5\ntPRk8bw4H22MOsPjtfKK5yUYpN9OJTERItNMYocPH0ZycjIWLlyI4uJieHp6wsLCov4DCZFC08Y7\nKEuKQye0MeosUhYXPA1gDK0oURAVkqlmceTIEcTHxyM+Ph7FxcXw8vLCsGE0GRlpPE0b76BoN08f\nQI83/wPX1zcwhpZdzyLqQqZkERUVhWHDhmHOnDlKmQeKNH+aNt5BoThOZJ0JALgSvh19x3ysknAI\nkUSm3lDHjh2Dr6+vSNlPP/2Ezz77TGGBvY56QzU/ERFx2LLlRK3xDiNbVHtF3NtOGH7kokjZc12g\nTQV9zol8KL3r7JAhQ7By5Uq8+eabAIC1a9fi9OnTiIyMlEsQsqBkQTRFfV2CBVWV0NIR7yByLy0O\nlrTWBJEjpXedDQ8Px+jRo9GqVStERkbi+vXrCA8Pl0sAhDQn9XYJ5jjJvUoYg6VyQiSkUWQelPfo\n0SO8+eabcHV1xY4dO8BxnKJjE0E1C6IJpE2B/u6IT3Dw5Hax8rKnj6FvYKSM0EgLpLSaRbt27USS\nQkVFBbKysnDw4EFwHIenT5/KJQhCmgtJXYIZOOC1RepS+hjB9cpj6CspLkKaqs5koczpPAhpDmp3\nCR7TdSnCH4jXMsCYeBdZQtRcnYPy8vLy6j1Bbm6u1G337t2Dp6cn+vTpAx6Ph71794rtw+fz0aFD\nBzg7O8PZ2RmrVkn4z0WIhggK8oad3RIwcGKJInb5hwA9SiUaqs42i/79+yM1NbXOE9S1T15eHvLy\n8uDk5ISCggIMHDgQFy9ehIHBq5lt+Hw+NmzYUG+DObVZEE1w1a49+twuEd9An12iAkprs3j9xi5J\n+/btpW4zNTWFqakpAMDExAR9+vRBSkoKPD09RfajJEA0XWXFC+i01hdbfCgz6Rjs3EapJCZC5KnO\nZFFVJb+pFzIyMnD16lUMHDhQpJzjOCQmJsLJyQleXl6YNWsW7Ozs5HZdQhSO4yT/R2IM9EkmzYVM\nEwk2VUlJCSZOnIiNGzeKrbTXv39/3Lt3D+fOnYO9vT1mz56tjJAIabJbMYcACV3Iy0uf0GMn0uw0\naPGjxnj58iX8/Pzg6+uLOXPm1LkvYwympqa4e/cuWrduLRoox2H58uXC1zweDzweTxEhE1I/aeOM\nKEkQFeLz+eDz+cLXK1asUM1KeQ3FGMMHH3wAExMTbNiwQeI+Dx8+ROfOncFxHMLDw7FlyxacOHFC\nPFBq4CYK0NDV+uKneWDY7jjxDfTZJGpI6dN91Hjy5AmePHkifG1lZVXn/gkJCdi9ezccHByEK+t9\n++23uHv3LgDA398fBw8exI8//ggdHR04ODhg/fr1DX0PhDRKg1fr4zi8PnPT6SlD4LU7QYFREqIe\nZKpZxMbGIiQkBOfOnYOenh4eP34Me3t7XLlyRRkxAqCaBWkaSTWIzZujJU7N4eOzFJGRK18V0CMn\noqGUXrNYs2YNdu3ahTFjxuDChQv4448/cObMGbkEQIiiSatB6OkVSNy/ZrW+8tInaG1gKLb9+qn9\n6OX1H8UES4iakqk3VF5eHqysrNC2bVs8e/YMkydPRkxMjKJjI0QupK33nZdXLHF/Pb0qgOMkJgow\nRomCtEgy1SyMjY1RUlICX19fTJgwAebm5ujdu7eiYyNELqSt921mZgZDQ9HV+kb3Ho0jURFi+1aU\nlaKVXluxckJaCpmSRVhYGPT09BAcHAw+n4+cnByMGzdO0bERJWtozyBNIW29b3NzAwQGjsSWLUvx\n4oU2+LErgHTRfYr0AKMyhlZKiJMQdSZTsnj06BFMTU2hr68PHo+HsrIyPHz4UGyAHdFcDe4ZpERN\nTWJ1rfft5zccBr/MxfAoCfObMQZaaYKQajL1hnJxccE///yDVq2qv1+Vl5dj6NChSElJUXiANag3\nlGJJW7RHrGeQkklKYnZ2S7Bpk0+dCeP1BDN4cFckJeWKr/ctoafT6YC34PXDcYW8H0KUSem9oQQC\ngTBRAEA/VZd1AAAgAElEQVSrVq1QUVEhlwCIepD2XL+mZ5Ay1b7RX7mSjsLCWSLbMzO/wZYtS6Um\nC2m1JJEEU0d3WC+5vAtCmheZekN5eXnhhx9+wMuXL1FRUYEffvgBb775pqJjI0ok7bm+np78JpOU\nRc2NPjp6FWJjQ1BYuB9AFADRUdN1JTFpvZ+2bDmB0sJciYkiMzGCxk0QUgeZksWcOXOQnJyM7t27\no3v37khOTsb8+fMVHRtRoppFe2qrfq4/UqlxSLrRA98AEJ0Cpq4kJq2WFBm1Cu1MuopvYAx2g30b\nGCkhLYtMj6EsLS3x66+/4uXLlwAAXV1dhQZFlK/m8UxNz6Dq5/pvKb1xW9qNHnhVk6hpnJbm9VrS\ndAt/7Lr/s4QzlqOb3QpsiohTeSM+Iequzgbu33//HdOmTcP69evB1aq6M8bAcRzmzZunlCABauBW\nNlV1o5XW0N6x4/vo27eXaOO0FLXbLBgkt01wePVZUnUjPiGKorQG7ufPnwOoXo+Ck9YgSDSGrAlA\nld1opXVz3bTpc5mv7ec3HEPe48EI34ptq50kaqiiEZ8QTVNnsvD39wcAjBgxAu7u7iLbaG4ozdKQ\nBCC9gVh6DyR5kcvjMI4TGx+x3ao7Prk7UeLuym7EJ0QjMRk4OTnJVKZIMoZKpPD2XsKqu/uI/vPx\nCRbb18NjucR9PTyWKz/whpAUNFDrvccy4GuRzXZ2i9nRo7GqjpwQhZDnfbPOmsU///yDxMRE5Ofn\nY8OGDcJnX/n5+ejYsaMSUhmRl4aMo1CXbrSyyr99BZ3s+omVX4/eh14j30dQRNxrj7aWQk/vDuzt\nDfB//zeRGrcJkUGdyaKiogIlJSWoqqpCSUmJsLxXr14ICgpSeHBEfhqSAOqaHkPtcBw6SSpnDL3+\n/VH80RYQGPiJ3JJEc51Ti5DaZJruIzs7GzY2NigrK4O+vr4y4hJDvaGaRvK0GV9j0ybJ7QEREXHY\nsuWE+PQYaiJuwXsYvv6gWLmg8iW0tBu0AGSTNHY6EkKUQZ73TanJoqKiQjjFR1paGr7++mtkZWUh\nPT0d58+fx/bt2/HDDz/IJQiZAqVk0WTqngBkpkYr16nrnFqEAErqOrtt2zY4OjrC3d0dq1atQmho\nKObMmQOgemLB6dOnyyUA0jiNefTh5zdcM5NDDTVKEjXUaU4tQhRJ6nQf/v7+2L9/PwAgNzcXjo6O\nwm3l5eVo06ZNvSe/d+8ePD090adPH/B4POzdu1fifosXL4atrS1cXFxw/fr1hr6HZiUiIg4+PsHg\n8ULg4xOMiIg4ifvUnj8pOnoVZs+OkrhvsyEhUcS866Ly+Zw0rTMAIY0mS5epkJAQ9tdffzEej8ey\ns7PZ/Pnz2cqVK+s9Ljc3l124cIExxlh+fj7r1q0be/r0qcg+Z8+eZUOHDmWFhYVs7969zM/PT+K5\nZAxVox09Gsvs7F7v2vm1WNfOhnSD1XhSusOqC8l/M+qOS9SDPO+bMk0kGBQUhIsXL6KwsBC+vr4w\nMjJCYGBgvceZmprCyckJAGBiYoI+ffqIrYFx9uxZTJgwAcbGxpg0aRLS09MlnapFWLr0D6mzpdbW\nEh593L90RmJt4nbScZXXJmrz8xuOTZt84OOzFB4eIfDxWSq10wAhmkymbiNGRkYICQlBSEhIoy+U\nkZGBq1evYuDAgSLlycnJmDZtmvB1p06dkJmZCTs7u0ZfSxNFRMQhPb1U4raaJFDTTnHx4j0AwQC8\nAby6KTWbRx8cBwsJxRFHY+Hnpn43YY1vCyJEBjIli8DAQLG5obp16wZfX1/07Nmz3uNLSkowceJE\nbNy4UWwpVsaYWGt9S5yHavPmaLx4YSVxm55elcQumkDNlOLD1XccRAPwA3zB+0l8hToOVQC0YDdb\nPZZ5lTcap0E0gUzJQltbG/Hx8Xj77bfBGMPRo0dRUlKCv/76C++9916dA/RevnyJd999F9OmTcPY\nsWPFtru5ueHatWvw8fEBUD063NbWVuK5atdseDweeDyeLOFrhOpHS16oTgCvEoKe3mcIDJwsdZ0H\nI6NJGDjwhEqmE5crjgNPUnGtif+UNT+VMqnz2udE8/D5fPD5fMWcXJaGjQEDBrDCwkLh68LCQjZg\nwABWXFzMBgwYIPU4gUDApk2bxubOnSt1n5oG7oKCArZnz54W28AtOn9RMAOWMyCYOTvPYIxp8HxN\n9ZHSgN1s3+9rWlRnBaJ08rxvylyzePr0KYyNjQEAT58+Bcdx6NChg3BBJEkSEhKwe/duODg4wNnZ\nGQDw7bff4u7duwCqu+cOHDgQ7u7ucHV1hbGxMXbv3t207KehRKfYqP5GaWf3NVaurB7Poqoumop6\nRMIEAnDa4g3yMQGj4PnDMbT2CZZ4XLNpl/lXS+isQJoHmZLF8uXL4enpiX79+oHjOFy+fBnff/89\nnj17hhEjRkg9zt3dHQKBoN7zr169GqtXr5Y96maovqm5VTFfk8IekXCc5CWJGIPnvz9q1PxUTUDj\nNIimqHduKIFAgH/++Qdubm5ISkoCx3Fwc3ODjo7y5t8BaLoPQPnTdch7Kov9G0Mxcd7XYuXZKSdh\n4/KmWHmzmZ6kDg2ds4uQhlDK3FC1OTk5IS0tTS4XbCxKFsrH44UgNjZErNzDIwR8vnh5naT0cIs4\nGtvib4otISkS1VDasqo1xowZg82bN+PDDz9E+/bt5XJhov7k8Ygkzq8vhh+7KlZe08vJp5n1bmoM\nGqdBNIFMyWLjxo14/vw55s2bJ5yinOM4PH36VKHBkWqq6off5HYDjoOkKGt3h6WGXEI0g0zJorRU\n8shiojg1CSInJx+3b3MoK/tJuE1Z/fDra3SXmsSkPHKqnSRqUEMuIRpC1j62SUlJLDQ0lDHG2J07\nd9jZs2fl1n9XFg0IVeOJTk6nnv3wJU6gZ7tI4piJ04GjFTrh3tGjsczbewnz8FjOvL2X0CR+hPxL\nnvdNmWoW3377LS5fvoy0tDQsWrQI7dq1w+effy42KSCRD9HR2urZD//1EeUMHHBbwo61usMC0msp\njUUjoAlRDpmSxZEjR5CQkAAXFxcAgLGxMSoqKhQaWEsmOlBLPfvh18Q4sNMOnM2fIbb9UcZFdLZz\nEClTREOupGlQmuO0IISomkzJwsLCQiQ5pKeno0ePHgoLqqUT7YXkjdfni1KHwWmtW1dW1ybyJWxk\nDJ2VFAeNgCZEOWRKFv7+/hgzZgwePXqEjz76CPHx8di2bZuiY2uxJE39oa8/EXZ2ZjA3N1D5pIHn\n+5kg6kqhWHl3u8XVg8mUGAuNgCZEOWQalAcAz58/x/HjxyEQCDBmzBjo6ekpOjYRLW1QntoO1JLS\n0+ktn2CVxEgjoAmRTukjuGs8fPgQL168EK43YWUlef0FRWhpyULtSFtjRA3+JmqbWAlRMaUniz/+\n+APBwcHQ1tZGq1athOWXL1+WSxCyoGShGi9fPIeufluxcv68d8Fbf1AFERFCZKX0ZOHg4ICIiAhY\nWlrK5aKNoanJQqNXQVPj2gQhpH5KnxuqY8eOMDAwkMsFWxJNHQNwdvNXcJu9Vqy88O4NdLQU7wWn\n0QmRECITmZJFr169MHz4cIwdOxaGhoYAqjPWvHnzFBqcptPIMQAcBzdJ5Yyho4RiTU2IhJCG0ZJl\npy5dumD8+PHQ0dFBaWkpSktLUVJSoujYNJ4mjQHINdSR/NipZnYOKaQnxBPyDpEQokIy1SxCQkIA\nAGVlZcJZZ0n9NGYMAMfBTFK5DM86NSkhEkIaT6aaRVpaGvz8/GBvbw8AuHjxIj7//HOFBtYcBAV5\nw85uiUhZ9ejrkSqK6DUc16jaRG0akxAJIU0iU7L45ptvsGbNGmF7haOjI2JjY2W6wMcff4wuXbqg\nX79+Erfz+Xx06NABzs7OcHZ2xqpV4st4aio/v+HYtMkHPj5L4eERAh+fpWoxWOz5kwKJSSJu8eQG\n93RS+4RICJELmR5DPXjwAH379hW+Li8vR5s2bWS6wEcffYTAwEBMnz5d6j4eHh4IDw+X6XyaRu1W\nQeM4SPzLMSZxoaL61LfmBSGkeZApWXh7eyMsLAwAcPfuXWzZsgVjx46V6QLDhg1DdnZ2nfto4vgJ\nTZO4ehaGLP5BrPzpo3to38miSedWu4RICJE7mR5DBQUF4cKFC6iqqsKoUaNgaGiIwMBAuQTAcRwS\nExPh5OSEefPmITMzUy7nJbVwnMREAcaanCgIIS1Dg+aGaqzs7GyMGTNG4vQgJSUl0NbWhq6uLnbt\n2oW///4bR48eFQ9UQ0dw10XRg9kEWhy0JP3KmtnvkRAimdJHcCtS7ZHhM2bMwJIlS1BeXo7WrVuL\n7VvThRcAeDweeDyeEiJUDIUPZuM4sWpjmQ6g/5ISBSHNFZ/PB5/PV8i5VV6zePjwITp37gyO4xAe\nHo4tW7bgxAnxAV3NrWbh4xOM6Gjxnl8+PksRGbmy8Sem+ZwIIf+S531TpjaLppg0aRKGDBmCGzdu\nwNLSEjt27MDWrVuxdetWAMDBgwfRr18/ODk54eDBg1i/fr2iQ1IL8h7M9iTvjsREkbh+DiUKQkiT\nyfwY6saNG4iKikJRUZFwPYtly5bVe9y+ffvq3D5r1izMmjVL1jCaDbkOZuM4dJBUzhiGNPxsNDEg\nIUSMTMni22+/RVJSElJTU/Hee+8hLCwMvr6+io6tWRNdOrVaQ9fWjv3yP/D47oBYednTx9A3MGpU\nXI1tS6EEQ0gzx2Tg6urKKisrmb29PWOMsfv377OhQ4fKcqjcyBiqRjl6NJb5+AQzD4/lzMcnmB09\nGiv7wa8m5RD910Te3kskntbHJ7jO92Fn97XI/nZ2Xzfs/RBC5E6e902ZahYcx0FbWxu9evXClStX\nYGNjg8ePHys2i7UAjRrMpuAG7Ma0pWjkVOyEkAaRqYF79OjRKCoqwmeffYYJEyagZ8+eCAgIUHRs\nBNWPd3x8gsHjhUhMFFmddeXagN2YthSaeZaQ5k+mmkVNQ/bIkSORnp6O8vJy6OnpKTQw8qr9ICPz\nW8k7MIZucr5mY9pSaOZZQpq/OpPFoUOHhP10OQnfasePH6+wwAiwfeNvyMjcLlb+xXAPfB/LV8g1\nGzMxoDwa6wkh6q3OQXkffvghOI5DcXExIiMj4ebmBo7jkJSUhFGjRuGvv/5SXqDNbFBevaS0TXBg\n8PAIAZ8fotx46hEREYctW07USjAjqb2CEBVT2nQfv/76KwDAx8cH58+fFy5+lJ6ejjlz5sglACIq\nNmgsPLaIT9euyz1DJaueXPz1xzvq0G2VZp4lpHmTqc0iNzcXFhavZic1NzdHbm6uwoJqsTgOHpKK\nwYB/vxy8/nhH4XNMEUIIZEwWM2fOxFtvvYUJEyaAMYbDhw/j008/VXRsLUcd3WEjIuLgU0f7AXVb\nJYQog0zJIjAwEEOHDsXx48fBcRy2bNkCZ2dnRcfWMkhIFAke3TCUfxtA/Y93qNsqIUQZZJ4bqn//\n/ujfv78iY9EIcmsfqKM2MbQBp6Fuq4QQZZBpUN7Jkyfh5eUFQ0NDGBgYwMDAAO3bt1d0bGqnpn0g\nOnoVYmNDEB29CrNnRyEiIk7mczzKuCgxUaQf/71Rg+uCgrxhZ7dEpKy6XWNkg89FCCHSyLSehaur\nKzZt2oTBgwdDS0vhs5pLpA5dZ5u8BoWCpuqgbqvNi7GxMYqKilQdBtEgRkZGEqdgUvpKea1atYKL\ni4vKEoW6aGz7AH/2OPA2h4mVCypfQku76YsVUrfV5qWoqEjlX4yIZpE0aFreZLpTDRs2DOPGjcN7\n770HQ0NDANXBtbQR3I1qH+A48CSVM6b4lacIIUROZEoWDx8+hKmpKc6cOSNS3tKSRYOmtaDlTQkh\nzYhS1uCWB3VoswBkbB+QkCj4/3EDb3+SkqIkmkxdPutEc0j7zMjzsyRTsqioqEBMTIzYsqo7duyQ\nSxCy0Ij/QFSbIHKgEZ91olaUkSxkemweHByMI0eO4O+//4aTkxOuXbuGLl261Hvcxx9/jC5duqBf\nv35S91m8eDFsbW3h4uKC69evyx65Gnlw9azERHEn+QQlCtJs2NjYoE2bNmjfvj2sra3h5eWFgwcP\nynRsdnY2tLS0IBAIFBwlURhZltPr378/EwgEwmVVHz9+zAYMGFDvcXFxcSw1NZX17dtX4vazZ8+y\noUOHssLCQrZ3717m5+cn9Vwyhqp8ClrelLRc0j7rR4/GMm/vJczDYznz9l7SqGVrm3IOGxsbdurU\nKcYYY+fOnWOLFy9m5ubmbP78+fUem5WVxTiOY5WVlQ2OmdRP2mdGnvdNmc5UkximTJnCoqOjWVZW\nFuvTp49MF8jKypKaLDZv3sw2btwofG1rays9UDW7ASd8N0dikhBUVak6NKLhJH3W5bHOeVPPUTtZ\n1Pjll1+YtrY2u3nzJjt69ChzcnJi7du3ZyNGjGC7du0S7mdpack4jmPt2rVj7dq1Y0lJSYwxxsLC\nwtiIESNY37592Y8//siePXsmPGbu3LmsZ8+ezNDQkA0YMIA9fPhQ5vfa0qhNsvj5559ZYWEhO3fu\nHPPw8GBvvPEG++OPP2S6QF3JYurUqSwqKkr42s3NjWVkZEgOVJ2ShYQk8UxXsfHJ41sl0QySPuve\n3kskVmB9fIJlPm9TzyEpWeTn5zMdHR22b98+xufz2ZUrV1hlZSWLjIxkBgYG7NatW4wxxrKzsxnH\ncayq1pep06dPMysrK3bixAl28+ZN9uabb7Lly5czxhg7evQo4/F4rKCggAkEApaamsqePn0q83tt\naZSRLGSedRaoHlnK5/Pl+QhMrPFFGYNLGuuMpx3c/53gTwRjaKPA69I05EQeE0YqYtJJExMT9OrV\nC/fv38f7778vLPfx8cHYsWMRFhaG+fPnS2xk/fvvvzFlyhSMGDECALBo0SLMnz8fISEhqKqqwtOn\nT5GVlYWOHTvSxKVqQKZkUVJSgujoaPzzzz8oLy8HUH1T37x5c5Mu7ubmhmvXrsHHxwcAkJ+fD1tb\nW6n7h4SECH/m8Xjg8XhNur6smEAATlsb7q+Vn938FdwCVyv8+jQNOZHHhJGKmHQyPz8f169fh6Wl\nJa5evYrvvvsOiYmJyMvLQ0VFRZ2zPiQmJmLRokXC1y4uLrh8+TJKSkrg5+eHe/fu4aOPPsKzZ88Q\nEBCA+fPnt/hZJOrD5/Pl+oW+NpmSxaeffgp9fX0MHjwYrVq1kromd0O5ublh3rx5mD59OqKiotC7\nd+8696+dLJTln0EWGHw2R3wDY3BTUgw0DTmRxzrnilgrPTy8elVHZ2dnzJ49G66uroiNjYWpqSmm\nTp0qrFFoa1d/VmvXMIYOHYqUlBS8++67AICUlBT069cPBgYGAIBZs2Zh1qxZSE9Ph5+fH+zt7eHn\n59foWFuC179Er1ixQm7nlilZXL16FZcuXWrwySdNmoTY2FgUFBTA0tISK1aswMuXLwEA/v7+GDhw\nINzd3eHq6gpjY2Ps3r27wddQlOdPCtDGsBMGv1Zekp8DA5OuSo2FpiEnNTXILXUshKWMc9Tc7FNT\nU3H48GHs2rULgYGB6NGjBx48eAATExN06NAB4eHhCA8Px7hx4wAAFhYW6Ny5M1JSUuDmVv01a+zY\nsfjoo48wcuRIWFlZYd26dXjnnXcAVH9D7tixI+zt7dGuXTtoaWkJkwhREVkaNn744Qe2atUqlpmZ\nyQoLC4X/lEnGUOXiTOjnYq2A8SN6SN2/KY3P0o6tXe7sPIOZms59rRfLYmrkbqaU+VlvCBsbG6av\nr88MDAyYpaUl4/F4bP/+/cLtJ06cYEOGDGEmJibs/fffZ19++SWbNm2acPv333/P+vTpwwwNDdnZ\ns2eZQCBgf/31F3vzzTdZnz592Pfffy/sDbVv3z7Ws2dP1q5dO+bs7MxWrlyp9PerSaR9ZuT5WZJp\nBPeuXbsQEBAAIyMjtGrVCkB1m8Xt2xIaexVEGaNaix9kwdBcQptJHdeV1PhsZ7cEmzb51PuNTdqx\nU6eaY/fuHJFyU9MZ6NpVDwYGnWga8maORnCThlLGCG6Z0k63bt1Ydna23DJUY8gYaqPFTBsmVpt4\ncPVsvcc1pTuitGM7dvxPk7tJEs2l6M86aX6kfWbk+VmSqWtB9+7doa+vL5/spIZyjHTA+z1e+Dru\nizEAYzCzH1jvsU1pfJZ2bGWl5N81NWgTQlRFpgZuIyMjODo6YsSIESLrWTS166w6YAIBOpZWNxSX\nawOVj/Iw3Lj+ea9qNKXxWdqxOjpljT4nIYQogkzJYtSoURg1ahSAV8/A1HnwXENwWlp4nJYEJqiC\neb8haN2AYyMi4pCf/xh6etPx4oUVAG8Aw2XujiitK+PUqR7YvVu+XRwlxb55czTKy3XQunUlgoK8\nqQ2EECJVg9azuH37dp2D5hRJ2Y1+9d1MJTVO6+kFoHfvl1i5crrMN15p62Mocl3tpjTKE8WjBm7S\nUGrTwB0TE8MGDhzIrK2tGWOMpaamsjFjxsit4UQWMoYqF7JMuCaPuXpURZNjbwmU+VknzYO0z4w8\nP0syNXCvW7cO4eHhMDIyAlA9WlOZ3WaVTfr0GieEr+tq2I6IiIOPTzB4vBD4+AQjIiJOofE2RERE\nHJKT70ncRg3ohBBpZGqzKC0tFVnsqKSkBO3bt1dYUKomORHEITn5Fni8ELRuXYl7925JPLakJF9t\nJ/2refxUXGwpcTs1oBNCpJGpZjF27Fhs3rwZlZWViIuLw2effYaJEycqOjaVEe+lFAcgCkVFfyA2\nNgTR0auQlWUAYIbIXqamc8FYeb21ElV5VWPyBrBEZFt1A/pIlcRFWraGrKL366+/YtiwYXKPYc+e\nPcIJTVUpJCQE06ZNU3UYEsmULAICAtC+fXvY2NhgzZo18PX1xWeffabo2FQmKMgbdna1b6bRAEQT\nAGO/ANADsBRACIClMDMrQfv2FhLPqQ6PeF7VmIYD8EFN7EZGk7BpU8PmCCItk42NDVq3bo3CwkKR\ncmdnZ2hpaeHu3bsqiqxppkyZgqioqCafR0tLq0mP6NW5l6lMj6H09fXx4Ycf4sMPP1RwOMpXV6+n\nmgnX4uOvQfKXnk6oThTV2rcPUetJ/0RjG/7vP2DgQJrqnMiG4zjY2tpi3759+OKLLwAAly9fRllZ\nmVrf6JSJNdOebPXWLP7++28MGTIEHTt2hImJCdzd3REWFqaM2BSu5hl+dPQq4eOl2bOjEBERBz+/\n4YiMXIkvv/SqY2oo0QSgp1cloVaiPo941Dk2ojmmTp2K3377Tfh6165dmD59ushN8vnz5/jxxx/R\nr18/eHt748iRI8JtjDHs3bsX9vb2cHJyQlycaAeQsrIybN++XTgr9YEDB6TegG/evImgoCBYWVlh\n9uzZuHVLclsiAKxevRrdu3dHx44dMWXKFMTHv5q1ofbjLUmPxXg8HrZv3w4AyM3NxZQpU9C1a1d0\n6tQJkyZNAgAMH179hcvR0REGBgY4cOAAiouLMXr0aHTu3BlvvPEGli1bhkePHgnPm5+fj4ULF8LU\n1BTvvvsuSktLRWJOTU3Fhx9+CBsbGyxduhQPHjyQ+v4Urc6axd9//41FixbB398fo0ePhkAgwNGj\nR7F48WIA1W0ZmkyWRYU2b44GY3NQ/Yy/9r4fA/hQ+Kpm0Jw8poFWFHWOjciGWyG/b+9seeO+AQ8a\nNAi///47rl+/jjfeeAP79+9HQkICgoODhfusWbMG8fHxOHToEO7fv48ZM2bAwMAAPB4PERERWLp0\nKXbt2oX27dvj008/FamVLFmyBLm5ufjzzz/x5MkTTJ06FYaGhhg5UvxLjbe3N2bOnIkLFy5g27Zt\n8Pb2RlZWlsS4u3fvjjNnzqBDhw746aefMHnyZNy7J7ln4Os4jhPGuGHDBpibmyMzMxPa2tpISUkB\nAMTFxUFLSwuXLl0Sjkd7/PgxZsyYgYMHDyInJwezZs3C5s2bsWrVKgDVj/j19fWRlpaGqKgozJo1\nC+PHjwdQnXA9PDywceNGrFu3DqtWrcLkyZMVtrhRfepMFt999x2+/fZbYfAA0LNnT3Tv3h3r1q3T\n+GQhy7xO1fvU3EyXAtAGUAVt7YdwdNwPA4PTYjddP7/hansDVufYiOaYNm0afvvtNwwfPhz29vYw\nNzcX2R4WFoa1a9eiR48e6NGjB6ZMmYLDhw+Dx+Ph2LFjmDJlCtzdq9ee/PTTT5GcnAygutZx+PBh\nxMXFwdKyutfejBkz8Pfff4sliwsXLqCiogJLllTXlhctWoTvv/8eFy5ckLgM64QJE4Q/z5kzBxs3\nbsT58+fh4uLSoPcuEAjw6NEjPHr0CNbW1hgyZIjUfY2NjYVrdNjZ2WHBggWYO3cuVq1ahcrKSpw8\neRKpqakwNTXFBx98IKy9AEBUVBT69euHTz75BAAQGhqKTp06oaCgACYmJg2KWR7qTBYPHz6UuDLV\nW2+9hfnz5yssKGWRpX3h1T6vnvEDgKPjLJw//z8FRkeIuMbWBuSJ4zhMmzYNw4YNQ1ZWltgjqJKS\nEly6dEnkJuzi4oKVK1cCAJKTk/H1118Lt9W+sV+/fh13796Fg4ODsEwgEKBbt25icSQkJKB///4i\nZa6urjhz5ozEZBEeHo5ff/0VSUlJKCsrQ2lpqVicsvj666/x3XffYfDgwbCxscHChQuFizy9jjGG\nJUuWID4+HpcvXwZjDKWlpWCMIT09HQKBQGRWDGdnZxQVFUl8f23atMEbb7yBf/75B2PGjGlQzPJQ\nZ5uFgYEBWrcWny2pdevWzWKchSzP8KXt83//13y7DhNSHysrK9ja2uL48eMiTx6A6vuGg4OD8PEM\nUL1kas0z/YEDB+LChQvCbampqcKfe/bsCQsLC1y7dg1FRUUoKirCkydPkJaWJhbD0KFDRY4FgPPn\nz0vsWvvs2TPMnDkTH3zwAa5fv47Hjx/D3NxcYltIp06doKuri7y8PABAZWUlLl++LNzesWNHhIaG\n4jnYm8oAABo5SURBVMGDB1i2bBmmTJkivMFraWmJnPPPP/9EREQEdu7ciYKCAhw6dAiMMTDG0KtX\nL2hpaSEzM1Pi78Ld3R3nz58XeQ+3bt2qsyajSHXWLNLT09GvXz+J22q/QU0lyzN8es6vWDShoeba\nvn07iouLoa+vj8pK0Vr62LFjsW7dOnTr1g0PHjzAvn37sGPHDgCAr68v5syZg7feegsGBgYij160\ntLQwceJEfPXVV1i8eDF69uyJrKws5OTkCJNNDWdnZ7Rq1QqhoaGYOXMmduzYAR0dHTg5OYnFWlJS\ngtLSUpiZmUEgEAhv9pK0bdsWgwYNws8//4ygoCD873//Q0lJiXD7gQMHMHjwYHTt2hVt27ZF27Zt\nhWuMu7i4ICUlBXZ2dgCABw8ewNDQECYmJrh58ybWrFkjPI+uri5GjBiBFStWYO3atTh58iTS0tKE\ntaiRI0di+vTp2LFjB0aPHo3Q0FAMGDAAHTt2lPlvJFd1zQWSlZVV5z9lqidUooFkmYOrJVLnz7qN\njQ07deqUWPnLly+ZlpYWu3PnDmOMsdLSUrZlyxbWp08fNmLECHb48GEmEAgYY4xVVVWx33//nfXu\n3Zs5Ojqy3377jWlpabGqqirGGGPPnj1jO3bsYB4eHqxDhw7M2dlZuHzrr7/+yoYNGya8bnp6Ops1\naxazsLBgX3zxBbt+/brU2Ddt2sR69OjBrKys2PLly5mnpyfbvn07Y4yxnTt3Mnd3d+G+iYmJzNvb\nm3Xt2pWFhISI7Ltw4UJmbm7O2rdvz7y8vNjBgweFx/3555/M1dWVGRoasgMHDrDi4mI2adIk1qlT\nJ9a/f3+2d+9ekff68OFDNn/+fNalSxc2fvx4saVoz507x6ZPn86srKzY4sWL2f379yW+N2mfGXl+\nluqcdZbJMBV5ffvExcXB398flZWVCAoKQmBgoMh2Pp+PsWPHCp/bvfvuuyK9KmrQTJzNj49PMKKj\nV0koX4rIyJUqiEg90Gdd+X7++WccPnwYx48fV3UojaKMWWfrfAw1bNgw8Hg8TJ48GT179hRWtSor\nK3Hjxg3s3bsXfD4fCQkJUs8xe/ZsbN26FdbW1vDx8cGkSZPEWvI9PDwQHh4uh7dDNElTVhkkRF4q\nKyuRlJQk7J1FJKuzgTs2NhYuLi5YsGABrK2tYW1tDSsrK1hbW2PBggVwdXUVGdjyuidPngCoHqxi\nbW0Nb29vnD17Vmw/+hbVMqnzaHfScvznP/9BUVERJk+erOpQ1FqdNQttbW288847wn7CT58+Bcdx\nMDAwkOnk586dQ69evYSv7e3tkZSUJNIdl+M4JCYmwsnJCV5eXpg1a5awcYg0b9JWCpTnioCE1Oev\nv/5SdQgaQaa5oWooorts//79ce/ePejq6mLXrl2YPXs2jh49KnHfkJAQ4c88Hg88Hk/u8RDloZ5m\nhMgXn89X2AjvBi2r2lBPnjwBj8cT9qkODAzEW2+9JXGgH1D9OMrU1BR3794VG99BjX6kpaDPOmko\nZTRwyzRFeWN16NABQHWPqOzsbJw4cQJubm4i+zx8+FD4Zo4cOQIHBweJAwGVSZ1XuiOEEFVo0GOo\nxvjvf/8Lf39/vHz5EkFBQTAxMcHWrVsBAP7+/jh48CB+/PFH6OjowMHBAevXr1d0SHWqmYlWHVe6\nI4QQVVHoYyh5UlbVnPr+E1Wjx1CkoTT+MZQmor7/hCiXspZVtbGxwalTp8TKAwICsGjRokadEwDi\n4+NFen02V5QsXkN9/wmRTpOXVa29JkWN2NhY3Lx5E6GhoY0+77Bhw3D9+vWmhqf2KFm8hlaTI0S6\n2suq1tDkZVXv3r2LvXv3amTsykbJ4jV+fsOxaZMPfHyWwsMjBD4+S7FpE/X9J6SGpi6rWltycjIG\nDx6MoKAgjBs3Dt9//73IzLlaWlr4/fff4eTkBHNzc2zcuBF5eXnw8fGBhYUFli9fLtyfz+cLF2oC\nqmtfP/30EwYPHgwrKyuEhITg5cuXAFDvMqthYWHw9PSEoaEhbG1tsXfvXpnej1LIbUpCBdOgUAlp\nkjo/67Wn6G3qv0awsbFhJ0+eZD179mTp6emssrKSWVhYsDt37jCO44Szzi5btox5enqyGzdusFOn\nTjEbGxsWExPDGGPsyJEjzNbWlsXHx7OLFy8yNzc3kZlY586dy95//32WlZXF0tLSWN++fVl0dDRj\nTHx2WGtra7Zq1SpWUFDAQkNDmY2NTZ2x18yYm5KSwpKSklhlZSVLSEhgVlZW7MSJE8J9OY5jI0eO\nZLdu3WKnT59m2trazMvLi8XGxrLMzEzWrVs3FhcXxxhjLCYmhllYWIhcx9HRkSUnJ7ObN28Kf2eM\nMVZYWMj++usvVlZWxjIyMpiPjw9bsmQJY4yxiooKZm1tzZKSkhhjjOXl5bGrV6/K9HeR9pmR532T\nahaEkAarWVb1xIkTUpdVXbRoEXr06AEvLy/hsqoARJZVdXBwwKeffiqsObB/l1Vdu3YtbGxs4Ojo\nKFxW9XW1l1Xt2LEjFi1ahJcvX4osrCSNi4sL3NzcoK2tjSFDhmDq1KkICwsT2ScgIADdu3eHp6cn\nbG1t4eTkhOHDh8PW1hYjRoyQ2FheY/r06RgwYADeeOMN+Pj44MSJEwBeLbOqp6cnXGa15rocx6Gi\nogIZGRl4/vw5unTpAnt7+3rfi7IofJwFIUSO1KBLraYuq1pbbm4u1qxZg8TERDx48ADl5eXo3bu3\nyD6Ojo7Cn7t06SL2OicnR+r5ay/AZGZmhoyMDOF7CQ4OlrjMqo6ODg4dOoS1a9ciMDAQ77zzDoKD\ngyW+d1WgmgUhpME0bVnV133zzTd4/vw5jh07hvv374vUbuSt9nkPHDggdZlVABg8eDAOHz6M7Oxs\n6OrqYuHChQqJqTEoWRBCGmX79u04ffo09PX1xbbVLKt68+ZN8Pl87Nu3D+PGjQNQvazqvn37kJCQ\ngEuXLkldVjU9PR0CgQCZmZlijeCA6LKqBQUFWLt2rdRlVV939+5dtGvXDm3atEFMTAx2795d7zG1\nb/qNTSx1LbP66NEjhIWF4dmzZ9DW1oaenp7MM3wrAyULQkij2NraijwGqt39dOHChRg3bhzGjx+P\nb775Bhs2bICHhweA6mQREhKCmTNnYvr06QgICBA5NiQkBJ6enggICICxsTHee+895OXlCa9Re9/I\nyEjk5OTA2dkZ9+7dQ2RkpEyxh4SE4PLly7C0tPz/9u48Jor7/QP4e20FW2ksqChFOUTlhl0EFgXk\nrOCJbT2ACjWoUQq2YoihmlbUVvFEPMGDYk0MtaU2lkpQAytoK5dGG88i4FGEcigiV8F9fn8Y5uvK\n4qKysPh7Xskk7JzveYD97MzOzAebN29GRESEwnqVXUr7/HRV8yubNywsDIaGhhg7dixCQkIQFhYm\nTJPL5YiPj4ehoSEsLCxQW1uLNWvWdGl/egI/7oMxDcN/6+xl8eM+GGOMaQRuLBhjjKnEjQVjjDGV\nuLFgjDGmEjcWjDHGVOLGgjHGmEr8uA/GNIyuri4/Mpu9FF1dXbVvQ+1HFjk5ObC0tMSYMWOwc+dO\npfN89dVXGDVqFMaNG/f/ohMRxl6ktrZWeAQEDzx0ZaitrVX736XaG4svv/wSSUlJOH36NHbv3o3q\n6mqF6fn5+cjNzUVhYSGio6MRHR2t7khqI5PJejtCl3DO7sU5u1dfyNkXMnY3tTYWdXV1AICJEyfC\n2NgYkyZNQl5ensI8eXl5mDVrFvT09BAUFIRr166pM5Ja9ZU/IM7ZvThn9+oLOftCxu6m1saioKBA\noSNzKysrnD9/XmGe/Px8hWe2Dx06FLdu3VJnLMYYYy+p16+Gaj/n9iz+co8xxjQMqdHDhw9JLBYL\nryMjIyk9PV1hnh07dtC2bduE16NGjVK6LjMzMwLAAw888MBDFwczM7Nuez9X66WzgwYNAvD0iigj\nIyOcOnUKq1evVphHKpVi+fLlCA0NRWZmZofeqtq19zTFGGOs56n9Povt27dj8eLFaG1txRdffIEh\nQ4YgKSkJALB48WI4OzvDzc0Njo6O0NPT61InJIwxxnpWn+nPgjHGWO/p9S+4VenKTX09ycTEBHZ2\ndpBIJHB2dgbwtIP6gIAAGBkZYebMmXj8+LEw/44dOzBmzBhYWVnh7NmzaskUFhaGYcOGwdbWVhj3\nKpmuXbsGBwcHjBo1CqtWreqRnLGxsRgxYgQkEgkkEgkyMjJ6Pefdu3fh5eUFa2treHp64siRIwA0\nr6ad5dS0mjY3N0MqlUIsFsPFxQXx8fEANKuenWXUtFq2e/LkCSQSCaZPnw6gh2rZbd9+qIlYLKYz\nZ85QWVkZmZubU1VVVa/mMTExoZqaGoVxGzdupMjISGpubqaIiAjavHkzERFVVlaSubk53b59m2Qy\nGUkkErVkysnJoQsXLpCNjc1rZZo8eTKlpqZSdXU1ubq6UkFBgdpzxsbG0tatWzvM25s579+/Txcv\nXiQioqqqKjI1NaVHjx5pXE07y6mJNW1oaCAioubmZrK2tqabN29qXD2VZdTEWhIRbd26lYKDg2n6\n9OlE1DP/7xp9ZNGVm/p6Az135i4/Px8LFiyAtrY2wsLChIx5eXnw9/eHkZERPDw8QESor6/v9jzu\n7u4dng3zMpnaP4XcuHEDc+fOxeDBg/Hxxx93e62V5QQ61rO3cw4fPhxisRgAMGTIEFhbW6OgoEDj\natpZTkDzavruu+8CAB4/foy2tjZoa2trXD2VZQQ0r5b37t3DiRMnsHDhQiFbT9RSoxuLrtzU19NE\nIhG8vb0xc+ZMHD9+HIBiTgsLC+Tn5wN4+ot69uouc3NzYZq6vUymvLw8FBcXQ19fXxjfk7XeuXMn\nXFxcsHHjRqExzc/P14icxcXFuHLlCpydnTW6pu05pVIpAM2rqVwuh729PYYNG4bIyEgYGRlpXD2V\nZQQ0r5ZRUVHYvHkz+vX739t3T9RSoxsLTXTu3DlcunQJGzZswPLly1FRUaH0k0dneuqGw9fN9DLL\nv47w8HCUlpYiMzMTt27dEq6UU7b9ns5ZX1+PuXPnIj4+Hjo6Ohpb02dzDhw4UCNr2q9fP1y6dAnF\nxcXYs2cPLl68qHH1VJZR02qZnp4OfX19SCQShXX3RC01urFwcnJSeArtlStX4OLi0ouJAAMDAwCA\npaUlZsyYgd9++w1OTk7CM62uXbsGJycnAE/vIbl69aqw7PXr14Vp6vaymUaPHo3Kykph/NWrV3uk\n1vr6+hCJRBg0aBAiIiJw7NgxjcjZ2tqKTz75BCEhIQgICACgmTVVllNTawo8vUBkypQpyMvL08h6\nPp9R02r5xx9/4Pjx4zA1NUVQUBCysrIQEhLSI7XU6Mbi2Zv6ysrKcOrUKeEwuzc0NjYKh6FVVVXI\nzMyEv78/pFIpkpOT0dTUhOTkZKHozs7OyMzMxJ07dyCTydCvXz+89957PZL1VTJZWFggNTUV1dXV\nOHbsWI/U+v79+wCAtrY2HDlyBFOmTOn1nESEBQsWwMbGBsuWLRPGa1pNO8upaTWtrq7Gw4cPAQA1\nNTU4efIkAgICNKqenWXUtFquX78ed+/eRWlpKVJTU+Ht7Y3Dhw/3TC1f4wv5HiGTycjCwoLMzMwo\nISGhV7OUlJSQvb092dvbk7e3Nx08eJCIiB49ekQzZsygkSNHUkBAANXX1wvLbN++nczMzMjS0pJy\ncnLUkiswMJAMDAxIS0uLRowYQcnJya+U6cqVKySRSMjExIRiYmLUlrN///40YsQIOnjwIIWEhJCt\nrS2NGzeOoqKiFK40662cubm5JBKJyN7ensRiMYnFYsrIyNC4mirLeeLECY2r6eXLl0kikZCdnR1N\nmjSJDh06RESv9n+jrpydZdS0Wj5LJpMJV0P1RC35pjzGGGMqafRpKMYYY5qBGwvGGGMqcWPBGGNM\nJW4sGGOMqcSNBWOMMZW4sWCMMaYSNxbsjZeZmYlLly71dowesWvXLjQ0NPR2DPYG4saCdcmDBw+w\ncOFCmJmZwcrKCi4uLvj11197O1YHU6dOxaNHj4TX2dnZOHXqFOzt7V9rvTo6Oq8bTe0SExPR2NiI\ngQMHdmn+9r5ZLly4ILyura196e12Vpv4+HgYGxtj6dKlL71OpnnU3q0qezMsWrQIY8eOxdmzZ2Fg\nYIDi4mK1NxZPnjzBW2+99VLL/P777wqvvby84OXl9dpZeuoBkC/S1taGt99W/i9LRNDW1saKFSu6\nvD6RSASZTAY9PT3h9avobLmoqCjo6emhsLDwldbLNAsfWTCVGhoaUFRUhPXr1wsPUhw9ejSio6MB\nPH2j2r9/Pz788EP4+vril19+AQDIZDL4+PggMDAQVlZWCr1x3bhxA+Hh4ZBKpYiIiEBNTQ0AwNPT\nE6tWrYKjoyMSEhKQnp4OFxcXSCQSfP7558In36amJmzZsgVSqRT29vbCA96e/XT8888/w9vbG97e\n3sL0srIyWFlZISIiAlZWVliyZAlaW1s77HN5eTkWLFgACwsLrF+/XmHaTz/9hGnTpsHd3R379u1T\nWrOCggKEhoZCKpUiJiYGLS0tQr64uDjY2dlh2rRpKC0tBfC0p7Zt27bBw8MDU6dOhUwmAwCkpKRg\n9uzZ8PX1hZ+fH1paWrB27VpYW1sjMDAQrq6uuHDhAkQiEdauXSvs+0cffYRx48Yp7HtXNTU1YcqU\nKThw4AC2bNki9FAZFRUFHx8fAEBWVhbmzZsnLLNu3TpYW1sjODhY4eiEHxDxBunWh5WwN9KPP/5I\n8+bN63R6dnY2LV++nORyOT1+/JgkEgm1tLRQdnY29e/fn65fv07Nzc1kY2NDd+/eJSKi6dOn0507\nd4iIaPfu3RQXF0dERJ6enhQUFEQtLS1ERPTgwQNhOxs3bqTExEQiIvr+++8pNDRU6N2sfb72ngxr\na2vJ3NycysvL6d69ezR27Fiqq6uj0tJSEolEdPr0aXry5An5+fnRmTNnOuzT0qVLadOmTSSXy+nr\nr78mHR0dIiIqLS2lOXPmUGtrK7W0tJCHhweVl5d3WN7Ly4sePnxIREQrVqyg1NRUId8333xDRETf\nfvstrVmzRtif9mefVVRUkLOzszBeV1eXSktLiYjo6NGjNGfOHGppaaGsrCwSiURUVFSksO9ERLW1\ntUREVFdX12kPjc/3+mhiYkJlZWXk6+tLhw8fJiKi8+fP0+zZs4mIyM3NjaRSKbW2tlJsbCzt27eP\niIhEIhEdOHCAiIgWLlwoPFeJiCglJYUiIyOVbp/1LXxkwVR6/jRDZGQkxGKx0Ad5Wloa0tPT4eDg\nADc3N9TV1QkdqTg7O8Pc3Bza2tqYMGECzp07h3///Re5ubmYMWMGJBIJEhMTce7cOWH9wcHB0NLS\nAvD06b6LFi2Cra0tkpOTcfLkSQBPjxrCw8OF3s3ef/99YXkiQkZGBiZNmgQDAwMYGhrC19dX6D/Z\n0NAQPj4+6NevHzw8PPDnn3922OfMzEyEhYVBJBIhLCxMGJ+Wlob8/Hw4OTlBKpWivLwcWVlZCssW\nFRXhr7/+gqenJyQSCdLT05GTkyNMDw0NBQB4e3sL205LS8P+/fshkUjg7++PyspKlJSUCPOZmJgA\nAE6ePInAwEBoaWnBy8sLxsbGSn9nqamp8PHxgaurK0pKSnD58mWl8z2LiBAQEICwsDDhqMHBwQFF\nRUWor6/HgAEDMH78eBQWFuLs2bNwd3cHALz99tv49NNPO+wTe7PwdxZMpcmTJyM6OhpEBJFIhF27\ndqGmpgaOjo4AnvYwtnLlSnz22WcKy8lkMoVuVLW0tNDS0gK5XI7Bgwfj4sWLSrf3wQcfCD9/9913\nmDhxIpKSknD8+HEkJCQAePrGRi84xSESiTp0DiMSiSASiRQaFi0tLYXO7Z+lbP1yuRzz58/H6tWr\nO922XC6HjY0NsrOzlU5vr0n//v3R3NwsLLN7925MnDhRYd7c3Fzh1F9XlZSUYO/evcL3ERKJRHj8\n9ouIRCK4ubkhIyMDQUFBQkZTU1OkpKRgwoQJsLOzQ1ZWFoqLi4We2bS1tTFgwIAO+8TeLHxkwVTS\n0dGBo6MjVq1ahfLycgBQuDwzODgYP/zwA6qqqgAAN2/eRGNjY6frGz58OExNTZGWlgYiQmtrq0IH\nLc++Sf/zzz8YPXo0mpubcejQIWH8rFmzhKt/ACi8GYpEIvj7++P06dOoqKgQPv1Pnjy5y+fQ/f39\ncejQIcjlcqSkpAjjAwMDkZaWhjt37gj52ve7nZOTEyorK4Wjq4aGBvz9998v3F5wcDCSkpKE/lLa\nG9Ln8/r5+eHo0aP477//cObMGdy+fbvDusrLyzF06FDo6ekJPTt21dq1a6Grq4uIiAhhnLu7O7Zs\n2QIPDw+4u7sjMTERDg4OXV4nezNwY8G65MCBA6isrISrqyucnZ0xf/58bNq0CQDg6uqK4OBgzJ49\nG7a2tggPD0dbW5vwSV6ZPXv2IDs7G2KxGBKJROHUxbPLrFy5EsuWLYO7uzvEYrEwLTAwEDY2NsL4\n9i+E2+nq6mLdunUICgrCvHnzsGHDBqHTl+czKcsYExODq1evwsrKCtra2sI8I0eORGxsLJYsWQI7\nOzvMmTNH6ZHJ4cOHsXfvXtjZ2WHChAm4ceNGh3merc+sWbPg7OwMPz8/2NjYCEcuz9dw2rRpMDc3\nh1gsxt69e2Ftbd3hVJSbmxuMjY1haWmJ7du3w9fXt8O2lWnfTkJCApqamhATEyOsr6KiAuPHj4e+\nvj7eeecd4RTU8/V70e+c9W3cnwVjfYhcLkdrayu0tbVRUFCAZcuWKXzf8zJMTU1RWFiIwYMHd3PK\n/0lJSUFRUZFwRRXru/jIgrE+pLGxEW5ubrC1tUVcXBy2bt36yusaOnQofH19hZvyult8fDzi4uKE\n7pFZ38ZHFowxxlTiIwvGGGMqcWPBGGNMJW4sGGOMqcSNBWOMMZW4sWCMMaYSNxaMMcZU+j+i9xS5\ni+as6wAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x10dc6bf90>"
]
}
],
"prompt_number": 15
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Tras verificar el c\u00f3digo, me di cuenta que antes estaba ajustando el modelo $y^{(\\lambda)}\\sim x$ en lugar de el modelo $y^\\lambda \\sim x$. Por ello hab\u00eda una discrepancia grande entre el modelo presentado en las l\u00e1minas y el modelo ajstado por mi. La gr\u00e1fica anterior muestra que tras la correcci\u00f3n, ambos modelos coinciden. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [],
"language": "python",
"metadata": {},
"outputs": []
}
],
"metadata": {}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment