{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", " \n", "
\n", "\n", "# Monod Example\n", "Author(s): Paul Miles | Date Created: June 18, 2018\n", "\n", "This example is from P.M. Berthouex and L.C. Brown: \"Statistics for Environmental Engineers\", CRC Press, 2002.\n", "\n", "We fit the Monod model\n", "$$y(t;q) = q_1 \\frac{t}{q_2 + t} + \\epsilon, \\quad \\epsilon\\sim N(0,\\sigma^2), \\quad q = [\\mu_{max}, K_x]$$\n", "to observations:\n", "\n", "| x (mg/L COD) | y (1/h) |\n", "|----------|:---------|\n", "| 28 | 0.053|\n", "| 55 | 0.060|\n", "| 83 | 0.112|\n", "| 110| 0.105|\n", "| 138| 0.099|\n", "| 225| 0.122|\n", "| 375| 0.125|" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Import required packages" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import scipy.optimize\n", "from pymcmcstat.MCMC import MCMC\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Initialize MCMC object, and create data structure" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# Initialize MCMC object\n", "mcstat = MCMC()\n", "# Next, create a data structure for the observations and control\n", "# variables. Typically one could make a structure |data| that\n", "# contains fields |xdata| and |ydata|.\n", "ndp = 7\n", "x = np.array([28, 55, 83, 110, 138, 225, 375]) # (mg / L COD)\n", "x = x.reshape(ndp, 1) # enforce column vector\n", "y = np.array([0.053, 0.060, 0.112, 0.105, 0.099, 0.122, 0.125]) # (1 / h)\n", "y = y.reshape(ndp, 1) # enforce column vector\n", "# data structure \n", "mcstat.data.add_data_set(x,y)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Define model and sum-of-squares function\n", "For the MCMC run we need the sum of squares function. For the plots we shall also need a function that returns the model. Both the model and the sum of squares functions are easy to write as follows" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "def modelfun(x, theta):\n", " return theta[0]*x/(theta[1] + x)\n", "\n", "def ssfun(theta,data):\n", " res = data.ydata[0] - modelfun(data.xdata[0], theta)\n", " return (res ** 2).sum(axis=0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Perform initial optimization study for estimating the error variance and computing the covariance matrix\n", "We define a residuals function and minimize it using scipy's optimize least-squares." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "def residuals(p, x, y):\n", " return y - modelfun(x, p)\n", "\n", "theta0, ssmin = scipy.optimize.leastsq(\n", " residuals,\n", " x0=[0.15, 100],\n", " args=(mcstat.data.xdata[0].reshape(ndp,),\n", " mcstat.data.ydata[0].reshape(ndp,)))\n", "n = mcstat.data.n[0] # number of data points in model\n", "p = len(theta0); # number of model parameters (dof)\n", "ssmin = ssfun(theta0, mcstat.data) # calculate the sum-of-squares error\n", "mse = ssmin/(n-p) # estimate for the error variance" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The Jacobian matrix of the model function is easy to calculate so we use it to produce estimate of the covariance of theta. This can be used as the initial proposal covariance for the MCMC simulation by assigning it to `qcov` in the `simulation_options` (see below)." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "tcov = [[2.44707212e-04 2.50115314e-01]\n", " [2.50115314e-01 3.20837579e+02]]\n" ] } ], "source": [ "J = np.array([[x/(theta0[1]+x)], [-theta0[0]*x/(theta0[1]+x)**2]])\n", "J = J.transpose()\n", "J = J.reshape(n,p)\n", "tcov = np.linalg.inv(np.dot(J.transpose(), J))*mse\n", "print('tcov = {}'.format(tcov))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Setup parameters, model settings, and simulation options\n", "- Parameters: \n", " - $\\mu_{max} \\in [0,\\infty]$\n", " - $K_x \\in [0,\\infty]$\n", "- Model Settings: \n", " - $\\sigma^2_0 = 0.01^2$ (initial error variance)\n", " - sum-of-squares function defined above\n", "- Simulation Options:\n", " - Number of simulation: 5000\n", " - Update error variance\n", " - Initial covariance matrix from least squares optimization" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "mcstat.parameters.add_model_parameter(\n", " name='$\\mu_{max}$',\n", " theta0=theta0[0],\n", " minimum=0)\n", "mcstat.parameters.add_model_parameter(\n", " name='$K_x$',\n", " theta0=theta0[1],\n", " minimum=0)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "mcstat.simulation_options.define_simulation_options(\n", " nsimu=5.0e3,\n", " updatesigma=1,\n", " qcov=tcov)\n", "mcstat.model_settings.define_model_settings(\n", " sos_function=ssfun,\n", " sigma2=0.01**2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Run Simulation" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Sampling these parameters:\n", " name start [ min, max] N( mu, sigma^2)\n", "$\\mu_{max}$: 0.15 [ 0.00e+00, inf] N( 0.00e+00, inf)\n", " $K_x$: 49.05 [ 0.00e+00, inf] N( 0.00e+00, inf)\n", " [-----------------100%-----------------] 5000 of 5000 complete in 1.1 sec" ] } ], "source": [ "mcstat.run_simulation()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Extract results:\n", "- Display chain statistics" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "---------------------\n", "name : mean std MC_err tau geweke\n", "$\\mu_{max}$: 0.1573 0.0341 0.0039 73.9652 0.9094\n", "$K_x$ : 68.1782 50.6119 6.0702 88.0618 0.7321\n", "---------------------\n" ] } ], "source": [ "results = mcstat.simulation_results.results\n", "names = results['names']\n", "chain = results['chain']\n", "s2chain = results['s2chain']\n", "names = results['names'] # parameter names\n", "mcstat.chainstats(chain, results)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Plot mean model response and compare with data" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# A point estimate of the model parameters can be calculated from the\n", "# mean of the |chain|. Here we plot the fitted model using the\n", "# posterior means of the parameters.\n", "xmod = np.linspace(1e-4,400)\n", "plt.figure(1)\n", "hmodel, = plt.plot(xmod,modelfun(xmod,np.mean(chain, 0)),\n", " '-k', label='model')\n", "hdata, = plt.plot(mcstat.data.xdata[0], mcstat.data.ydata[0],\n", " 's', label='data');\n", "plt.xlim([0, 400])\n", "plt.xlabel('x (mg/L COD)')\n", "plt.ylabel('y (1/h)')\n", "plt.legend(handles=[hdata, hmodel]);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Visualize parameter chains and pairwise correlation:\n", "- `plot_chain_panel`: plots the sampling history of the simulation\n", "- `plot_pairwise_correlation_panel`: chains plotted against each other" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAFmCAYAAABpxD1nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi40LCBodHRwOi8vbWF0cGxvdGxpYi5vcmcv7US4rQAAIABJREFUeJzt3X+cHFWZ7/HP0yOEKyFRrhDAIYCSC4qsrBBIAkqEZYMIu+t6gejiJu4Qfr78QUxIIqwKiEkEQRTMQpiXsMjCRuGiwF2j4LgrZCABQXBRFxECcyEBBRLI8kMyz/3jdDnVNdU9/buru7/v16tfM1NVXX3KSD1d5znnOebuiIiIRHKtboCIiGSLAoOIiBRQYBARkQIKDCIiUkCBQURECigwiIhIAQUGEREpoMAgIiIFFBhERKSAAoOIiBRQYBARkQKZCQxm9nYz+46Z/cHMXjGzh83soNh+M7PzzeyZ/P47zGxK4hw7mtn1ZrbZzF40s34zG9/8qxERaV9vanUDAMzsrcDdwADwIeA5YArwQuyws4FPA3OAx4ELgNVm9m53fzV/zPXArsBRwDbAt4GrgI+X2Q4DdgNeqvGSRESy5CWvpGKqu7f8BSwDflZivwHPAAti2yYCrwKz83+/C3DgoNgxRwPDwG5ltuPt+XPopZdeenXSa0Il9+RMPDEAf0X49v9d4HDg/wHfcveV+f17AbsAd0RvcPdNZnYvMB24Mf/zRXe/L3beOwiB4RDg/yQ/1MzGAeOS25966ikmTJhQj+sSEWmZzZs3s/vuu1f8vqwEhncApwOXAF8BpgLfMLPX3f1aQlAA2Jh438bYvl2AZ+M73f0NM3s+dkzSEuCLyY0TJkxQYBCRrpWV5HMO+Lm7f97dH3D3q4CVwGkN/tylhC6p6NXb4M8TEcm8rASGZ4BHEtt+BUzO/74h/3NS4phJsX0bgJ3jO83sTcCOsWMKuPtr7r45eqGks4hIZgLD3cA+iW3/C1if//1xws39yGinmU0g5A4G85sGgbeY2YGxcxxBuMZ7G9BmEZGOlJUcw6XAGjP7PLAKOBg4Jf/C3d3Mvg6ca2aPMjJc9WnglvwxvzKzHwIrzew0wnDVy4Eb3f3pZl+QiEi7ykRgcPd1ZvYRQp//Fwg3/s+6+/Wxw74KbE+Yl/AW4C7g6NgcBoC/IwSDOwmjkW4izH0QEZEyWSVzHjpdvntq06ZNmzQqSUTa3ubNm5k4cSLAxHwetSxZyTGIiEhGKDCIiGTY0BAMDISfzaLAICKSUf39sMcecMQR4Wd/f3M+VzmGGOUYRCQrhoZCMBgeHtnW0wNPPAG9ZU7FVY5BRKSDPPpoYVAA2LoVfvvbxn+2AoOISAaNL7KSzPbbN/6zFRhERDLo5ZfTt2/Z0vjPVmAQEcmgKVMgl7hD9/TA3ns3/rMVGEREMqi3F666KgQDCD+vvLL8xHMtNCopRqOSRCRrhoZCwnnvvSsPCtWOSspErSQREUnX29ucp4Q4dSWJiEgBBQYRESmgwCAi0gStqHlULQUGEZEGa1XNo2ppVFKMRiWJSL3Vo+ZRtVQrSUQkg1pZ86haCgwiIg3UyhnM1VJgEBFpoN5eWL58JDg0cwZztRQYREQaqL8fFi0K3UlmsHQp9PW1ulWlKfkco+SziNRTKxPPoOSziEjmtGPiGRQYREQaph0Tz6DAICLSMK0snV0L5RhilGMQkUZIK509NBS6mqZMaVygUI5BRCSjenth5syRAJD1Ehl6YojRE4OINFozRyrpiUFEpA20w0glBQYRkSIaUSq7HUYqKTCIiDA6CDQqD9AOI5WUY4hRjkGkO/X3wymnhC6eXC7UNorKWETqnQdIG6lUb9XmGBQYYhQYRLpPWjI4lxudB4DwRDFzZtOaVjMln0VEqpCWDI4K3sVlLQ/QSAoMItLViiWDly9PzwO009rN1VJgEJGuViwZvHBhyCkMDISffX3Zn5hWL5nIMZjZl4AvJjb/xt33ze/fDvgaMBsYB6wGznD3jbFzTAZWAB8EXgauBZa4+xsVtEM5BpEuNVYyuNUltKtRbY7hTY1rUsX+E/iL2N/xG/qlwIeB44FNwOXAzcChAGbWA9wObABmALsC/wz8Efh8oxsuIu0lrU5Rb2/pG3ypiWlZDQzVylJX0hvuviH2+j2AmU0E+oD57v4Td78f+CQww8ym5d/7l8C7gZPc/UF3/zfgH4EzzWzbFlyLiGRUtd1B1UxMa9d8RJYCwxQze9rMfmdm1+e7hgAOBLYB7ogOdPdfA08C0/ObpgMPx7uWCN1NE4D9Gt90EWkHQ0Mj8xUg/Dz11PJu3JVOTGvnfERWAsO9wFzgaOB0YC/gZ2a2A7AL8Lq7v5h4z8b8PvI/N6bsJ3bMKGY2zswmRC9gh5quQkQyrdY6RX19oxPSaWoJQFmQiRxDvusn8pCZ3QusB04AXmngRy9hdNJbRDpU1B2UTCBXMj9hrFwEtH8+IitPDAXyTwf/BexNSChva2ZvSRw2Kb+P/M9JKfuJHZNmKTAx9mqDfzIRqVaz6hS1Q6G8UjIZGMxsPPBO4BngfsLooiNj+/cBJgOD+U2DwP5mtnPsNEcBm4FHin2Ou7/m7pujF/BSXS9EREpqRXK23O6gWrRDobxSsjKP4WLgVkL30W7AecABwLvd/TkzWwEcQ8hDbAa+CeDuM/Lv7wEeBJ4GzibkFa4Drnb3soerah6DSPMkC9dddVVjbtKt1IxCeaW0dRE9M7sR+ADwP4HngLuAc9z9sfz+aILbxyic4LYhdo49CBPcZgJbCBPcFmuCm0j2tONksXbU1hPc3H32GPtfBc7Mv4ods57wVCEiGdfuydlOl8kcg4h0tnZPznY6BQYRabp2T852ukzkGLJCOQaR5mp1crbTtXWOQUS6UzmTxaT51JUkIiIFFBhERKSAAoOIiBRQYBCRlmnX9Qo6nQKDiLREO69X0Ok0XDVGw1VFmkMlMZqj2uGqemIQkaardcEcaSwFBhFpuvHjwaxwm0piZIcCg4g0VX8/TJsG8V5slcTIFs18FpGmSa6FDKGY3uAgTJ3aunZJIT0xiEjTpOUWhodhy5bWtEfSKTCISNOUKretOQ3ZocAgIk1TrNz26tWa05AlmscQo3kMIs0RL7cNmtPQKCq7LSJtI15ue2BAy3xmjbqSRLpUVvr0tcxn9igwiHShLNUp0jKf2aMcQ4xyDNINslqnSMt81p9yDCJSllJ1ilp5Q9Yyn9mhriSRLpPWp5/Lwfbbt6Y9kj0KDCJdZGgoPDEsWzbSpw/hCWLaNM0fkECBQaRLxBPOixeHV9zwMJx6autHKUnrKTCIdIFk8brhYfjKV0YfpzURBBQYRLrCmjWjE85pAxJzOc0fEAUGkY7X3w+zZ5d37Pz5GhkkmsdQQPMYpNOkzVkoJgtzGaS+tOaziIySNmchzVizjbNSPkOaQ4FBpIOlzVkwGxmqmsvBggXhSaGvL/0cWSqfIc2hrqQYdSVJJ+rvD8NQt24NgWDZMvjYx8orP5HV8hlSHpXEEJFUfX3wwguwaFG4wS9eDDvuWPwJIS6r5TOksdSVJNLhhoZGggJUNpFNJbG7kwKDSIcr9a1/LCqJ3Z0yGRjMbLGZuZl9PbZtOzO7wsz+YGYvm9lNZjYp8b7JZna7mf23mT1rZheZmbrLpKvV+q2/ry/kFAYGSieppXNkLjCY2VTgVOChxK5LgeOA44HDgd2Am2Pv6wFuB7YFZgBzgLnA+Q1vtEiG1eNbf28vzJypJ4VukalRSWY2Hvg5cAZwLvCgu3/WzCYCzwEfd/fv5Y/dF/gVMN3d7zGzDwG3Abu5+8b8MacBy4Gd3P31Mj5fo5KkY2khnO7TKRPcrgBud/c7EtsPBLYB/rTd3X8NPAlMz2+aDjwcBYW81cAEYL+0DzOzcWY2IXoBO9TnMkSyIT4xTd/6pVyZCQxmNht4H7AkZfcuwOvu/mJi+8b8vuiYjSn7iR2TtATYFHtpXqd0DE1Mk2plIjCY2e7AZcDfufurTfzopcDE2EvfpaQjpJXZ1loLUq5MBAZCV9HOwM/N7A0ze4OQYP50/veNwLZm9pbE+yYBG/K/b8j/ndxP7JgC7v6au2+OXsBLdbgWkZarZYiqSFYCw53A/sABsdd9wPWx3/8IHBm9wcz2ASYDg/lNg8D+ZrZz7LxHAZuBRxrcfpG6qrVonSamSS0yERjc/SV3/2X8BWwB/pD/exPQD1xiZh80swOBbwOD7n5P/jQ/IgSA68zsvWY2C/gycIW7v9aCyxKp2NAQLFwIkyfXlhvQxDSpRaaGq8aZ2U/JD1fN/70d8DXgY8A4woijM9x9Q+w9ewArgJmEwHItsNjd3yjzMzVcVVqmv78wLxCppWidhqh2t2qHq2Y2MLSCAoO0ylgL6gwMhKGmIpXolHkMIl2p1II6yg1IsykwiDRYOYnktGQxhG3KDUizKTCINNDFF5c3ySwtWbxgAaxfX13ROi3FKbVQjiFGOQapp4sugrPPLtw2ViK5HsnieBI7lwsBRxVRu5OSz3WgwCCVGBoKuYEpU0bfxIeGYPfd09/XyESyluKUOC3tKdJEad/KZ80aCRRr1qS/z6yxieSxZjwXC2QicQoMIhVKq0M0b1646UeBYt689PeeeurITbnUE0e1oiR28olh3To48kh1L0l5lHwWKaJYAjftW7l7YaBYuTIEijgzOOec8HujKp+mJbFPPrn6NZ+lOykwiKQodeMuNrQ0bngYPve5whv0ypXhxt3oyqfRUpwLFoRzX3llCFxxKqgnpSj5HKPks0B5Cdz+/nAz37o1BAn3wptvdDyMHmU0MBACTlI9k9JjzaRWQro7aOazSJ2UU7I6+lY+MBDmGqxcmV6wLm3VtGZUPh1rJrUmzUkpCgwiCcW6iq64ovDv+E0/HiieeKJ0YrcZlU+LBZ9Vq8Zun4i6kmLUlSSRc8+FCy8cvX3tWpg6tT6f0ejKp/Hurij4KCB0F81jEKmDaAhpseTy3XfXLzBEXU2N0tcX5lao7LZUSoFBJC85aS3NoYc2t021anTwkc6kHIMI6UNIk/MQ5syp39OCSJbpiUGE4pPWVqyAV18NTwqVBoW0mc2NmO0sUm8KDCIULyVx7LHV3cDTailB8aqnChiSJRqVFKNRSd2tXqN4ik2QGx5OnwS3erXKZEtjqOx2HSgwSC1DSKNv/c89ByeeWN57Vq2C2bNVJlsaQ8NVReqg2lE8ya4js9FPB2lPDPHie5GtW+Gyy8JCPyKtoFFJIjUqNqIpGvIadUullc2YMSN9aOwll6j6qbSOnhika9Ur4Zs2oml4OHQT7bRTYbdU2oSz+fPD2tDJ9//2t+pOktbQE4N0pXhZ7cmTYeHC6r+hF6tLNH366AJ6aUX1PvOZ0XMm0orqFVsfQqTeFBik6yS7ftzDN/ZqF8yptSheb2/x6qyRRi3sI5JGo5JiNCqpOxRbDwFqGxFUa1G8Yu8vZ30IkTQalSRSprTJbJFo3YVqbrjFRjSVm8so9v5S60MoMEgjqCtJuk5vLyxblr4vlyu+YE4lffzRsVEXVS1dQM1Y2EckToFButJBB6Vvnz8//Vt4JX388WMXLqx9bedmLOwjEqccQ4xyDN2jkn77YscODsLLL48ukldqrWWofm3nRi/sI51Haz6LVCD5LTyXg6VLK+vjP+SQ0U8QpdZahtq6gNKGuoo0ggKDtL1qx/f39YVcQ5SIXrw4vYuo2BrQ0cN2vIuo2LGgLiBpHwoM0taK9f2XEyyGhmDRorFzAGlPF0nxUULJfMA554QSF4ODqpoq7UHDVaVtpdUoOvVUeOGFkRt+qTLWlQwDja+ffPPN8M1vjj7f9tuPPnbduvAkopLa0k6UfI5R8rm9FJuolrbgTiVJ5VITx0oll5NJ5WonpmnRHqmXtk4+m9npZvaQmW3OvwbN7EOx/duZ2RVm9gcze9nMbjKzSYlzTDaz283sv83sWTO7yMz0RNTB0vrz0yauRU8BSdUMAy2WXE6b/1DqiaQYlb6QLKg5MJjZZ+rQjiFgMXAgcBDwE+D7ZrZffv+lwHHA8cDhwG7AzbE29AC3A9sCM4A5wFzg/Dq0TTIq7cYeJZPjSo0E6usL3+AHBkIO4B3vGJ1jiOcriiWXly0rHLI6MADjx1fWlmJdYyqaJ03n7jW9gKuBK4Ge/N/vBm6ow3mfB/qAicDrwP+O7dsXcGBa/u8PAVuBSbFjTgM2AdtW8JkTAN+0aZNL+3jqKfeBgfDT3f3qq917etwh/Lz66rHPcfXV7rlceE8uN/KetO3x8+dy7hddVPw8c+aU35af/CQcl3wNDFT5P4x0vU2bNnn+XjnBK7j/1iXHYGZnAcfkb8R7Asvc/XtVnquH8GRwLfDnwC7AncBb3f3F2HHrga+7+6Vmdj7wV+5+QGz/XsDvgPe5+wNFPmscMC62aQdgSDmG9lfJZLBSE9imTUvPEcDo85c6z5YtY7dFxfKk3lqWYzCzqcD7gbcSbuQfrSYomNn+ZvYy8BrwT8BH3P0RQmB4PR4U8jbm95H/uTFlP7Fj0iwhBLPopYf2DhAlb6Mum7GGrRbLBdx1V+lRS8nJZsXOs2VLeRPTVPpCsqIeyedLgX9y94OA2cAtZnZoFef5DXAAcAiwArjWzN5dh/aVspTQVRW99J9gm7voorDwTrQAT/R7qURusSJ1hx1WWY6gHsXu4jmPJ57Q0FZpjbIDg5mdmrbd3Q9z9x/lf18HHAt8tdKGuPvr7v5bd7/f3ZcAvwA+A2wAtjWztyTeMim/j/zPSSn7iR2T9pmvufvm6AW8VGm7JTsuvhjOPntkRnLUSw+lE7lp39SXLg11kJYvL/8bfG8vfOIThdtOOqnyb/wqfSGtVnaOId/Nc6S731vimJ3d/Vkz287dX62pYWY/AZ4kBIfngI+5+035ffsAvwamu/s9+aGttwG7uvuz+WNOAS4Cdnb318r8TM1jaFPlFK+D0gXsorzEffeNBBizEBymTlWOQNpPMxbqORe4yczeF91848zsvcD3gT0rDQpmthT4N0Ig2AH4ODATmOXum8ysH7jEzJ4HNgPfBAbd/Z78KX4EPAJcZ2ZnE/IKXwauKDcoSHsbq3gdjN2tE928jzii8Klj0SJ48smxb+5aUEc6RdldSe7+deDfCcGhIKCY2XHAXcDaKtuxM/DPhDzDncBUQlD4cX7/WYQngpuA/yB0D/1trG1bCV1YW4FB4Dv5832hyvZImylVvA7CvrPOGvs8a9aMBIWIexhZVE0btKCOtKNKk88nA+OBb0QbzOxzhMlml7j7CdU0wt373H1Pdx/n7ju7+1/EggLu/qq7n+nuO7r79u7+t+6+IXGO9e5+jLu/2d13cvcF7v5GNe2R9rN6deEN3Swkop96ChYsCNui1dTqNZs4WahPo4qkU1Q8jyE/P2AdoWvpQEK3zyfdfVX9m9dcyjG0p7S+/VwO1q8Pv1fS7z80FEYyxf+ziM4VP76/f2SWcrI4nhbUkaxo+DwGM7vazE4njPY5GbgcmAUc1glBQdpXWt/+8HC4OVdar6i3F1auLCyxPX9+4TFjla7QqCJpd5V0JU0BlgFrgFWEada/BA4zs0PNbPsGtE9kTKX69qvp94/mEixYEJ4ckl1Ql11WeXE8kXZSSfL5cHefCOwD/D1hYtu2wHnAz4BNZvZIQ1opUkKpvv1a+v0vuWT0PIh16+BrXxt9rJLM0kkqLkvt7o8CjwI3RtvyeYeDCCUxRJouvjhOsm+/2L74ugdQuAZCqTIZaWm5s85S15F0jrqsV+DujwOPA9+tx/lEKhG/wRebvBY9PUTiyWOzsM19JJH8Z3+Wfp6oayqZ6P5MPYrPi2REJhbqEalWNQvbJJPHydIZp5wC11+f/t7x40d3TV11lZ4WpLNoac8YDVdtD9ETwvjxxctil7pRF1sSdCzxc2tIqrSDZpTEEGm55PyBakpQpHUHjSWXK0xaJ7umRDqJupKkpZKzh8c6Njl/IClt7eWk5EilKMdQyo03qgS2dA8FBmmZSvMD5RTKK7dnNL7uwY03lj62pwemTy/vvCKdQIFBWqKahe/Hjx/7vO7lTzSLZijvtVfxY8xU70i6jwKDtESlpSogLJwzlnK6kqo5r0g3UWCQlhirVEVa7mGs0toQnhhWr66sLXfeWfp8Yz3JiHQaBQZpiVKlKorlHpLvSVPpjXxoKCzjWYrqIEm30TyGGM1jaL7kfIBylsccGgoL58yeXTwZXWoJz+RxY81p0PKc0q40j0HaUnI+QDnLY/b2wvHHw+bNhQnsSCUF7caa06DFdqQbqStJMqWSMtl9fWEBnQULqqueGs2gXrZsdPdULhfO+8QTmr8g3UddSTHqSsqG/v6QJ9i6deRGP2tWYfXTpEpLVCRnUH/0o3DTTSN/L18+siSoSLuqtitJgSFGgSE74jf61asLb+LLlsFBBxUPEmnnigeVtDxGUi4H99wDU6fW75pEmk2BoQ4UGLKn1E08udZy8n2PPgr33w+LFhWuzfyOd5RXRM8sLPOpriRpVwoMdaDAkD1jjRpKGzEU7yZKO35wEA45pLzyGRqRJO2s2sCg5LNk2liT2pJzDNatg3nzincTbd0abvRJZunF9DSHQbqRAoNk2liT2uIjlvr7w/oMpZ4EenoKF+aJuMO3vjU6OGgtZ+lGCgySefFKqBddlD40NVmUL010/IwZ6UNijz025BSqGfoq0kmUY4hRjqE9pA1NLZaLiEYxTZ1aeHzakNgoyazV2aRTKPlcBwoM7Stt9FIuFwLA88/D+98/euipAoB0OgWGOlBgaG/Jp4BDDoE1a0b2z5kD11zTsuaJNJ0CQx0oMLS/6Cng5ZfhuONG71+7tvSkteRkOJF2puGqInnucMcd6ftuv734+5Llvhcu1DoM0p0UGCTTkgv2rFsHl1wSfibFb+yXXZZ+vl12Kf458fkPw8Nw8cUwefLYa1GLdBqV3ZZMiXflJGskTZtWPGdQznBVCENS0z7zqqvS5z9EC//MmqWuJekeCgzScsXqGsUnog0PFwYFgGuvhTPPDDmDtHUc0px7bmECulT5jEhyPQiRTqeuJGmpePfPwoWFXTnljIu4++7wM610RlqJi2uvHemGKvcpQ7OfpdsoMEjLlHtjLuWxx8LP3t6whkIUHHI5OPzw9PdEwaScpwzNfpZulInAYGZLzGydmb1kZs+a2S1mtk/imO3M7Aoz+4OZvWxmN5nZpMQxk83sdjP77/x5LjIzdZdl1Jo1tQUFgMsvD91D/f0j3VAQnjZ++tP09xx6aPg5VoE+reAm3SoTgQE4HLgCmAYcBWwD/MjMto8dcylwHHB8/vjdgJujnWbWA9wObAvMAOYAc4HzG998qVR/P8yeXZ9zXXghnHxyYZAp1g01Z87IPIaoQF9alxPApZfWp30i7SaTE9zMbCfgWeBwd/8PM5sIPAd83N2/lz9mX+BXwHR3v8fMPgTcBuzm7hvzx5wGLAd2cvfXy/hcTXBrgnJWUEtjVl7eIc2ZZxYGhXLbMjAAM2dW95kirdZpE9wm5n8+n/95IOEp4k/Tltz918CTwPT8punAw1FQyFsNTAD2a2hrpSLl9O2nlb++994wdLQae+yRPuP5ssuKt0VJZ+lWmQsMZpYDvg7c7e6/zG/eBXjd3V9MHL4xvy86ZmPKfmLHJD9rnJlNiF7ADjVfQJdLTkhLM1bfPoQAkCx/PXVqyCcU6/qJtqftX7JkdJuGhuBrX0s/Vy6npLN0r8wFBkKu4T1AnXqgS1oCbIq9VAChBsmSEmkzhqM5C8uXF198J5eDc84ZWYMhngB+9NHi3Unu4b2nnDJ6X9pKbMXOdeKJsH69ks7SvTKVYzCzy4G/Bj7g7o/Hth8B3Am8Nf7UYGbrga+7+6Vmdj7wV+5+QGz/XsDvgPe5+wMpnzcOGBfbtAMwpBxD5dL66pPrJccnk0XrJDz2WPhmHpesghqfDQ1j5yeiZTpLtaVYm3O5EBT0pCCdoK1zDBZcDnwEOCIeFPLuB/4IHBl7zz7AZGAwv2kQ2N/Mdo697yhgM/BI2ue6+2vuvjl6AS/V5YK6UFreIP4tPTlnYXg4DC9NBgWA73xnpNsn+RRy441wzDGl2+IORx019kpsyWVDe3rC3woK0u0y8cRgZt8CPk54WvhNbNcmd38lf8wK4BjCENTNwDcB3H1Gfn8P8CDwNHA2Ia9wHXC1u3++zHZoVFKVxnpiKLbCWjGrVsH06dWNXoo+e3AQtmwZeyEeLdgjnaraJ4asTP46Pf/zp4ntnwSuyf9+FjAM3ETo/lkNnBEd6O5bzexYYAXh6WELcC3whUY1utsl1y646qrRy2VGN9oo4VzuTX72bJg/v/oJcFu3hqBQzlDT3l4FBJG4TDwxZIWeGMqXzBdcdVVI1pb69h1fYa2cIBGNXBrruBUrwjyFtKcV0MI70r20glsdKDCUp1i30eBgWDlt/PjwM+1mHAWO7bcPS2/W4/9+a9fCQw8VPq0sWwYbN4a1G+LBa9YsBQrpHu3elSRtpFiiOXmjjz9JROLdNitXFj5BxMtsV+K228IiO088EYLOunVw9tmF5xoeLkx+m4XP15BUkdH0xBCjJ4byVFLSIm2YaPJc0RPEqlWhPtHWrZWXv4g/EUyeXN57zeDJJ/XkIJ2rrYerSntJDvMsJW1iWfJcjz0WVme7+OLqggKEIHXqqfDlL5f/XvfQ/SUihRQYpCp9feGmWqw8RVx8feZkyYy0NRnGurEX+8ytW0PASjNWCQ4RGaH/XKQsaTWQXn65vG/nUZ2itJIZ5S7JGeeeHhyiPEXSCSfAPfeMfk8uF+ZKiEghBQYZU7EaSOUUw4PwTf6220bPfD7llDDXoJynjqSoLlL0+dFIpGR7enpCobypU0OyWbOcRcam5HOMks+jjTWjOT6foZRa1lIAeP/74Wc/G7191SrYaaeReRPxuRLRJLv4yCPNcpZuouGq0hClaiAlb6zRN/+0AFDr94+0oNDTE7qC4u3o6wsjk4rd/DXLWWRsCgxSUlopi2gBm2Ti2D3su+EG+P3v4Ywz0s9ZD8UK44Fu/iK1Uo5BRoknmtMqkEY35GJPEzuR9DpHAAASd0lEQVTtBMcd17iRQJdeWrhGg4jUlwKDFEhLNPf1pS+ak5Z8jp4mKpnrUImeHjj00BCUSq0SJyLVU2DoUmnDT9PWTDj11JEnh5kzC7tokjf/aPGd6JgooCxaVJ829/TASSeFyXClVokTkdoox9CFilVGrSTRHOnrg+efDzf/aPEddzjooPBEccMNYRnPauy/P3zjG6FcxpYt4ee0aaMD16xZyimI1JOGq8Z0w3DVUsNPofi+Z56BW2+FXXcN+YPoRlyqblKtQ1RvvRWOPXbk71WrwnrMSQMD5a27INJtVCtJyjLWU0Fa19D8+XDwwXDBBWGk0eTJI104pWYu1xIU9tgDDjhgpLurvz8s3pMU5TREpH70xBDT7U8M0VPARReNdAkVk8vB+vXh92qX3xxLNEy22JNH2gQ2ERmhJwYpS6nhpxACx+LFY3/bHx4eecr4xCca09b4/Ig0N9ygoCDSCEo+d6FSs4MrKWr34x+H9193XWPaWUo061lE6k9PDF0qbfgplF8YD+ArXwnrH5QTSOr9zX54GFavru85RSRQYJBRcxrmzSv/vVdeWd5xb3978SqqM2ZUXmHVfWSORdqcDBGpngJDl4vPdJ48ObzKvdlX4vzzw8+0AHDvveF13HGVnXPrVrjssvSS4CJSPY1KiumGUUlxlazdXC/JgnyRFSvCnIW09nzqU7DjjnDeeaPPBaVHWIl0M41KkrJFXS9r1jQ2KLzznaO3Ffu800+Hc88dPY/iq18Ns5+/9CW4+urC0VTz5xefkyEi1dMTQ0w3PDEky2G4175WQj2tXRtmVxdbTyG+0A6MPSdDpJvpiaED1TupmlYkz6y6pTWrUc5op9tvhwcfDOUvHnxw9P74aKqx5mSISHX0xBCTpSeGYoXuajEwEJK0SStWwJlnNrZb6ayzwvm/8Y3STyh77QWPPz7y94wZcPfdpc+t5TpF0lX7xKDAEJOVwFBO2Yp6nvdf/gWefHKkQmrWJIvpiUh51JXUQUoVuqtFWvmKrVtDxdIsBIVp09K3//CHzW2HSLdTSYwMKrXOci2GhoqXr2hlUDALo4/23Td9LsPRRze/TSLdTE8MGVRuUrXS5HQldZCaackSWLAgdBfNmFG4b8YMdSOJNJtyDDFZyTFESiVVq0lOt2JCWzmS+ZNrroFbboG/+RuYO7eFDRNpc0o+10HWAkMxtSSnFy6Eiy9uaPOqEq3C1ojRWCLdSsnnLlJJcjrZ3XTCCY1vX6Wi/EnaPIuoUJ6INI8CQxtKK40dv7nGl8OMF5ibO7f4yJ9WiedPGjUaS0Qqk5nAYGYfMLNbzexpM3Mz+5vEfjOz883sGTN7xczuMLMpiWN2NLPrzWyzmb1oZv1mNr65V9J4acnppUtDpdHJk0Mg2H33UD47/u372muzk1/I5cLs5ieeGOkqKhXwRKR5MhMYgO2BXwBnFtl/NvBp4DTgEGALsNrMtosdcz2wH3AUcCzwAeCqRjW4lfr6wk11YACWLQvLcV58ceGs4qymj3p6QmA7/vjCnIhKXIhkQyaTz2bmwEfc/Zb83wY8DXzN3S/Ob5sIbATmuvuNZvYu4BFgqrvflz/maOD/Ar3u/nQZn9sWyee4rI40SvPtb8Oee45dukIlLkTqo9OTz3sBuwB3RBvcfRNwLxCt/DsdeDEKCnl3AMOEJ4yOlNW5CWn+8z/TlxNNKrbsqIg0R7sEhl3yPzcmtm+M7dsFeDa+093fAJ6PHVPAzMaZ2YToBexQvyY3RnKUUSVrNLfaJZdohJFIO2iTW0rDLAE2xV6Zvm0lRxn194dv1YsXt7pl5RkehsHBVrdCRMbSLoFhQ/7npMT2SbF9G4Cd4zvN7E3AjrFjkpYCE2OvzHZeFBvjf8IJ8JWvtLZtlZg9W+syi2RduwSGxwk39yOjDfmun0OA6DvoIPAWMzsw9r4jCNd4b9pJ3f01d98cvYCXGtH4eig2xv+7321Ne6qlSWsi2ZeZ6qr5+QbxEet7mdkBwPPu/qSZfR0418weJQSKCwgjlW4BcPdfmdkPgZVmdhqwDXA5cGM5I5KyLq3iqll2h6SWEk1aU3JZJJuy9MRwEPBA/gVwSf738/N/fxX4JmFewjpgPHC0u78aO8ffAb8G7iQMU70LOKXhLW+CtDH+S5a0tk3l0qQ1kfaSyXkMrdIO8xiSY/znzg0zmrPs+OPh5pvDk0I0aU2F8UQaT9VV66AdAkOaz342lMPIqp6eMBppyxZNWhNppmoDQ2ZyDFK+oaGQjB4/Hh5/HHbeeez3tNLWrSEozJzZ6paISDkUGDIouvFPmVJ6gZ52oZyCSHvJUvK56w0NhYV0ogqp0SS2+P54xdSsMgsvUCE8kXakwJAR0azmeIXU5Jj/W29tj+Gpp8TGgWU9iInIaEo+x7Qq+TxWhdRo2cu994bHHmtas+qm3GVHRaS+Or26akcrVSE1lwsB4dxz2zMogFZhE2k3CgwZMGXKSJ980rx5sGYNXHhhc9tUT0o+i7QXBYYM6O2F5cvT9115JZx4YnPbU6tjjtEqbCLtTMNVM2LhwvDUsGhR+ydsV6/WhDaRdqYnhgxZsAC+//1Wt6J28QltCgoi7UeBIWP+679a3YKx5XIhiK1dC6tWjc6PKKcg0t7UlZQx739/q1tQWi4H99wDU6eGv6dOhc2bw3yLeJE8PSmItC8FhozJcleSWSj9HQWFSF8fzJpVWPVVRNqXAkOGnHNOtpfp/MEP4Nhj0/f19iogiHQKBYYWiRfKA/jyl0MXTFbNmVM8KIhIZ1FgaIF4hdR2WZ5zv/1a3QIRaRaNSmqyoaHCstntEBQAzj47FPgTkc6nwNBkpeoiZd2iRSOVXkWkcykwNNmUKWHIZ9bceis89RScdFLxY4aHVQxPpBtk8BbV+ebNa3ULRvT0wNVXh8Ryby9cdx0ceGDxYzVxTaTzKfncRFlclvPkk8MchMjQEDzwwOjjNHFNpHvoiaHBhoZC2YgVK7K5LOeVVxYuIVosB3LDDWEim4h0Pq3gFlPvFdz6+0MwaIf/iaNV1mD0anJagU2kPWkFt4yJhqW2Q1CAkVXWentD2QutpyDSvZRjaJAsDks94QTYZhu4/vrR+6IlREG1j0S6nQJDg0TDUrMUHNJKZEeWLy8MAKp9JNK91JXUIFGXTNbmLLiH4BC1K5eDr341rK8gIgIKDA3V1wfr14cZw1niHpLiAwOhfQsXtrpFIpIlGpUUU+9RSQBz58K119blVHX31FPqLhLpZBqVlBFDQ+Gb+NAQ3HZb84NCsuuqpwcOOyz92Ntua3x7RKT9KPlcR60upz1nTlgaNLnM5vr1cNddo4/fsKG57ROR9qCupJhaupKGhkZPDGuWqVPhiitGltwcGiocarpuHRx88Oj3rV07eplOEekc1XYl6YmhTpo9b8EMvvAF+PCHR9/ck0NNp04NTxPxbq05cxQURCSdAkMdDA3Bc881b95C1EVUSe2ia66BM8+Eu++GQw9VUBCR4hQYapTMK9QzOJjB5z8Pb3tbuJnvumtts5GnTlVAEJGxdVxgMLMzgYXALsAvgE+5+9pGfFbaMp1mYYbxww/DBRdUdr6DD4bLLy8dADS8VEQaraMCg5mdCFwCnAbcC3wWWG1m+7j7s/X+vLS8wvAw7LQTHHfc2IHh1lvhlVfC79Onjy5JISLSCh0VGID5wEp3/zaAmZ0GfBj4B2BZvT8srR5StMpZb+/ohG/cnDlh1TQRkazpmMBgZtsCBwJLo23uPmxmdwDTi7xnHDAutmmHSj4zqoeUnDcQfduPJ3z33jt0PW3YkD6SSEQkKzomMABvA3qAjYntG4F9i7xnCfDFWj50rBLVSviKSLvppMBQjaWEnERkB2Co0pOoRLWIdJJOCgy/B7YCkxLbJwGpxR/c/TXgtehvK7ZYgYhIF+mYInru/jpwP3BktM3Mcvm/B1vVLhGRdtNJTwwQuoWuNbP7gLWE4arbA99uaatERNpIRwUGd/9XM9sJOJ8wwe1B4Gh3TyakRUSkiI4KDADufjlweavbISLSrjomxyAiIvWhwCAiIgU6riupHjZvLns9CxGRzKr2XqYV3GLM7O1UMcFNRCTjKlrBTYEhxsIMt92Al2o8VTSDurcO52on3Xrd0L3X3q3XDe117S95BTd7dSXF5P+H+3+1nic2g/qlSqJ0u+vW64buvfZuvW7o7GtX8llERAooMIiISAEFhsZ4DTiPWIG+LtGt1w3de+3det3Qwdeu5LOIiBTQE4OIiBRQYBARkQIKDCIiUkCBoUpmdqaZPWFmr5rZvWZ2cIlj9zOzm/LHu5l9tpltracKr3uemf3MzF7Iv+4odXzWVXjtf2tm95nZi2a2xcweNLNPNLO99VLJdSfeNzv///dbGt3GRqnw33xu/nrjr1eb2d56UWCogpmdSFgU6DzgfcAvgNVmtnORt7wZ+B2wmCLLjLaDKq57JnAD8EFgOvAU8KN86ZG2UsW1Pw9cSLjuPyMsFvVtM5vVhObWTRXXHb1vT+Bi4GcNbmLDVHntm4FdY689Gt3OhnB3vSp8AfcCl8f+zhFmTC8u471PAJ9t9TU0+7rzx/cQ/sP5+1ZfS7OvPf+enwMXtPpaGn3d+X/nu4E+4BrgllZfRzOuHZgLvNjqdtfjpSeGCpnZtsCBwB3RNncfzv89vVXtarQ6XfebgW0I36bbRq3XbsGRwD7AfzSqnfVWw3V/AXjW3fsb28LGqeHax5vZejN7ysy+b2b7NbipDaHAULm3Eb4RJZcL3UhYTrRT1eO6lwNPE/uPrU1Ude1mNtHMXgZeB24HPuXuP25YK+uv4us2s8MITwrzGtu0hqvm3/w3wD8Afw2cRLi/rjGz3kY1slFURE+awswWA7OBme7elgm5KrwEHACMB44ELjGz37n7T1vaqgYxsx2A64B57v77Vren2dx9EBiM/jazNcCvgFOBf2xVu6qhwFC53wNbgUmJ7ZNo48RyGaq+bjNbQEi8/4W7P9SY5jVUVdee73r4bf7PB83sXcAS4KcNaGMjVHrd7wT2BG6NVR7NAZjZG8A+7v5YQ1pafzX/d+7ufzSzB4C969y2hlNXUoXc/XXgfsI3QADMLJf/e7DY+9pdtddtZmcTvi0d7e73NbqdjVDHf/McMK6+rWucKq7718D+hKek6PUDYCD/+1MNbnLd1OPf3Mx6CP97PNOINjZUq7Pf7fgCTgReBeYA7wKuBF4AJuX3/zOwNHb8toz8h/I0cFH+971bfS0Nvu5FhAJjHyX0y0av8a2+liZc+xLgKOAd+eM/B/wROLnV19LI6055/zW076ikSv/NvwD8Zf7f/H2EodqvAO9u9bVU+lJXUhXc/V/NbCfgfMKN7kHCN+IoUTUZGI69ZTfggdjfC/KvfyeM9W8LVVz36YSg+L3Eqc4DvtTY1tZXFde+PfAtwuperxC+TZ/k7v/avFbXrorr7hhVXPtbgZX5Y18gPHHMcPdHmtfq+lB1VRERKaAcg4iIFFBgEBGRAgoMIiJSQIFBREQKKDCIiEgBBQYRESmgwCAiIgUUGEREpIACg4iIFFBgEBGRAgoMIjUws8PM7I9mtl1s2575heDbc71f6XoKDCK1OQD4lRcuPvTnwAvuvr5FbRKpiQKDSG3eS2HlXAjB4hctaItIXSgwiNTmAEI55rg/T9km0jYUGESqlF+h6z2MfmJ4H7HAYGY/MLPLzOweM/uNmR1sZt83s/Vmdkb+mJPMbK2ZPWxmt5vZuPz2u83skPzv/WZ2VnOuTrqZAoNI9fYBtiOsygeAmU0H3k7hE8P+wEPuPg24k7CC30nAB4FP5o/5N3c/2N33z59vZn77BcBiM5sPDLv7pY27HJFAgUGkegfkf37KzKaY2YcIyz1CWLkOM9uBsCBWf+x933D3lwADNpuZAfPMbJ2Z/YKwFOqrAO7+Q8JKYR8Gzmj4FYmgwCBSiwOA1YQ1fh8GLgS+CGwGPp0/Zj9gXew9+wP35n9/T/59c4F9gQ+4+3sJy0I+AmBmU4EdgU3u/scGXovIn2jNZ5HqvRdY5+7nJrb/S+z3/YGHYn/3uvtQbN/DhOBxt7u/YmZnAm929+fM7O3A1cARwE1m9h53/2VDrkQkRk8MItV7L+HGXsqfAoOZ7Q48ldj3MHAdcLaZ3QPsBTxsZv8D+C7wKXd/HFgK/GN9my+Szty91W0QaTtmtgvwDLCfuz/S6vaI1JMCg4iIFFBXkoiIFFBgEBGRAgoMIiJSQIFBREQKKDCIiEgBBQYRESmgwCAiIgUUGEREpIACg4iIFFBgEBGRAgoMIiJS4P8D+24VgW5jNXAAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# plot chain panel\n", "mcstat.mcmcplot.plot_chain_panel(chain, names)\n", "plt.savefig('monod_chain_panel.eps',\n", " format='eps',\n", " dpi=500,\n", " bbox_inches='tight')\n", "# The |'pairs'| options makes pairwise scatterplots of the columns of\n", "# the |chain|.\n", "mcstat.mcmcplot.plot_pairwise_correlation_panel(chain, names, figsizeinches=(4,4));\n", "plt.savefig('monod_pairwise.eps',\n", " format='eps',\n", " dpi=500,\n", " bbox_inches='tight')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Generate credible intervals\n", "Instead of just a point estimate of the fit, we should also study the predictive posterior distribution of the model. We can calculate the model fit for a randomly selected subset of the chain and calculate the predictive envelope of the model. The grey areas in the plot correspond to 50%, 90%, 95%, and 99% posterior regions." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Generating credible/prediction intervals:\n", "\n", "\n", "Interval generation complete\n", "\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "def predmodelfun(data, theta):\n", " return modelfun(data.xdata[0], theta)\n", "\n", "mcstat.PI.setup_prediction_interval_calculation(\n", " results=results,\n", " data=mcstat.data,\n", " modelfunction=predmodelfun)\n", "mcstat.PI.generate_prediction_intervals(\n", " nsample=500,\n", " calc_pred_int=False)\n", "# plot prediction intervals\n", "mcstat.PI.plot_prediction_intervals(adddata=True)\n", "plt.xlabel('x (mg/L COD)', Fontsize=20)\n", "plt.xticks(Fontsize=20)\n", "plt.ylabel('y (1/h)', Fontsize=20)\n", "plt.yticks(Fontsize=20)\n", "plt.title('Predictive envelopes of the model', Fontsize=20)\n", "plt.savefig('monod_ci.png',\n", " format='png',\n", " dpi=500,\n", " bbox_inches='tight')" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.8" }, "latex_envs": { "LaTeX_envs_menu_present": true, "autoclose": false, "autocomplete": true, "bibliofile": "biblio.bib", "cite_by": "apalike", "current_citInitial": 1, "eqLabelWithNumbers": true, "eqNumInitial": 1, "hotkeys": { "equation": "Ctrl-E", "itemize": "Ctrl-I" }, "labels_anchors": false, "latex_user_defs": false, "report_style_numbering": true, "user_envs_cfg": false }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": true, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": false, "toc_position": {}, "toc_section_display": true, "toc_window_display": true } }, "nbformat": 4, "nbformat_minor": 2 }