3_Introduction-to-PyMC.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:6d349ea64b8e4405dbdbcc86fe0989fa84f5bb12221f16825b5fd46491d78f8e"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# Introduction to PyMC\n",
      "\n",
      "While Markov chain Monte Carlo is a powerful method for fitting Bayesian models, they can be difficult to apply generally, as most commercial statistical analysis packages do not implement it. This has been an impediment to the growth of Bayesian methods. PyMC is a python module that implements Bayesian statistical models and fitting algorithms, including MCMC. Its flexibility and extensibility make it applicable to a large suite of problems. Along with core sampling functionality, PyMC includes methods for summarizing\n",
      "output, plotting, goodness-of-fit and convergence diagnostics.\n",
      "\n",
      "PyMC 2.3 provides functionalities to make Bayesian analysis as painless as possible. Here is a short list of some of its features:\n",
      "\n",
      "-   Fits Bayesian statistical models with Markov chain Monte Carlo and\n",
      "    other algorithms.\n",
      "-   Includes a large suite of well-documented statistical distributions.\n",
      "-   Uses NumPy for numerics wherever possible.\n",
      "-   Includes a module for modeling Gaussian processes.\n",
      "-   Sampling loops can be paused and tuned manually, or saved and\n",
      "    restarted later.\n",
      "-   Creates summaries including tables and plots.\n",
      "-   Traces can be saved to the disk as plain text, Python pickles,\n",
      "    SQLite or MySQL database, or hdf5 archives.\n",
      "-   Several convergence diagnostics are available.\n",
      "-   Extensible: easily incorporates custom step methods and unusual\n",
      "    probability distributions.\n",
      "-   MCMC loops can be embedded in larger programs, and results can be\n",
      "    analyzed with the full power of Python.\n",
      "    \n",
      "Before we dig into PyMC in detail, let's look at a simple hierachical linear model of a house's price as a function of age, to give you a flavor for what Bayesian models look like when implemented in PyMC. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "\n",
      "# Data\n",
      "age = np.array([13, 14, 14,12, 9, 15, 10, 14, 9, 14, 13, 12, 9, 10, 15, 11, 15, \n",
      "                11, 7, 13, 13, 10, 9, 6, 11, 15, 13, 10, 9, 9, 15, 14, 14, 10, 14, 11, 13, 14, 10])\n",
      "price = np.array([2950, 2300, 3900, 2800, 5000, 2999, 3950, 2995, 4500, 2800, 1990, \n",
      "                  3500, 5100, 3900, 2900, 4950, 2000, 3400, 8999, 4000, 2950, 3250, \n",
      "                  3950, 4600, 4500, 1600, 3900, 4200, 6500, 3500, 2999, 2600, 3250, \n",
      "                  2500, 2400, 3990, 4600, 450,4700])/1000.\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pymc import Normal, Gamma, deterministic, MCMC, Matplot\n",
      "\n",
      "# Constant priors for parameters\n",
      "a = Normal('a', 0, 0.0001)\n",
      "b = Normal('b', 0, 0.0001)\n",
      "\n",
      "# Precision of normal distribution of prices\n",
      "tau = Gamma('tau', alpha=0.1, beta=0.1)\n",
      "\n",
      "@deterministic\n",
      "def mu(x=age, a=a, b=b):\n",
      "    # Linear age-price model\n",
      "    return a + b*x\n",
      "\n",
      "# Sampling distribution of prices\n",
      "p = Normal('p', mu, tau, value=price, observed=True)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This example will generate 10000 posterior samples, thinned by a factor\n",
      "of 2, with the first half discarded as burn-in. The sample is stored in\n",
      "a Python serialization (pickle) database."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M = MCMC(locals(), db='pickle')\n",
      "M.sample(iter=20000, burn=10000)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--------         21%                  ] 4368 of 20000 complete in 0.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [---------------- 43%                  ] 8660 of 20000 complete in 1.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------63%----              ] 12675 of 20000 complete in 1.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------82%-----------       ] 16557 of 20000 complete in 2.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------100%-----------------] 20000 of 20000 complete in 2.4 sec"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "Matplot.plot(b)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Plotting b\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAl4AAAFwCAYAAABpb3VdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhLmcSi4AAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x102452cd0>"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Example: Coal mining disasters\n",
      "\n",
      "Recall the earlier example of estimating a changepoint in the time series of UK coal mining disasters."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import matplotlib.pyplot as plt\n",
      "from pymc.examples.disaster_model import disasters_array\n",
      "\n",
      "plt.figure(figsize=(12.5, 3.5))\n",
      "n_count_data = len(disasters_array)\n",
      "plt.bar(np.arange(1851, 1962), disasters_array, color=\"#348ABD\")\n",
      "plt.xlabel(\"Year\")\n",
      "plt.ylabel(\"Disasters\")\n",
      "plt.title(\"UK coal mining disasters, 1851-1962\")\n",
      "plt.xlim(1851, 1962);"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAuAAAAD9CAYAAAD9CcJlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107a102d0>"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We represent our conceptual model formally as a statistical model:\n",
      "\n",
      "$$\\begin{array}{ccc}  \n",
      "(y_t | \\tau, \\lambda_1, \\lambda_2) \\sim\\text{Poisson}\\left(r_t\\right), & r_t=\\left\\{\n",
      "\\begin{array}{lll}             \n",
      "\\lambda_1 &\\text{if}& t< \\tau\\\\ \n",
      "\\lambda_2 &\\text{if}& t\\ge \\tau             \n",
      "\\end{array}\\right.,&t\\in[t_l,t_h]\\\\         \n",
      "\\tau \\sim \\text{DiscreteUniform}(t_l, t_h)\\\\         \n",
      "\\lambda_1\\sim \\text{Exponential}(a)\\\\         \n",
      "\\lambda_2\\sim \\text{Exponential}(b)     \n",
      "\\end{array}$$\n",
      "\n",
      "Because we have defined $y$ by its dependence on $\\tau$, $\\lambda_1$ and $\\lambda_2$, the\n",
      "latter three are known as the *parents* of $y$ and $D$ is called their\n",
      "*child*. Similarly, the parents of $\\tau$ are $t_l$ and $t_h$, and $\\tau$ is\n",
      "the child of $t_l$ and $t_h$."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## PyMC Variables\n",
      "\n",
      "At the model-specification stage (before the data are observed), $y$,\n",
      "$\\tau$, $\\lambda_1$, and $\\lambda_2$ are all random variables. Recall from the discussion of subjective probability that the Bayesian interpretation of probability is ***epistemic***, meaning random variable $x$'s probability distribution $p(x)$ represents our knowledge and uncertainty about $x$'s value, rather than some random generative process. Candidate\n",
      "values of $x$ for which $p(x)$ is high are relatively more probable, given what we know. Random variables are represented in PyMC by the classes `Stochastic` and `Deterministic`.\n",
      "\n",
      "The only `Deterministic` in the model is $r$. If we knew the values of\n",
      "$r$'s parents, we could compute the value of $r$\n",
      "exactly. A `Deterministic` like $r$ is defined by a mathematical\n",
      "function that returns its value given values for its parents.\n",
      "`Deterministic` variables are sometimes called the ***systemic*** part of\n",
      "the model. The nomenclature is a bit confusing, because these objects\n",
      "usually represent random variables; since the parents of $r$ are random,\n",
      "$r$ is random also.\n",
      "\n",
      "On the other hand, even if the values of the parents of variables\n",
      "`switchpoint`, `disasters` (before observing the data), `early_mean`\n",
      "or `late_mean` were known, we would still be uncertain of their values.\n",
      "These variables are characterized by probability distributions that\n",
      "express how plausible their candidate values are, given values for their\n",
      "parents. The `Stochastic` class represents these variables.\n",
      "\n",
      "First, we represent the unknown switchpoint as a discrete uniform random variable:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pymc import DiscreteUniform, Exponential, Poisson, deterministic\n",
      "\n",
      "switchpoint = DiscreteUniform('switchpoint', lower=0, upper=110)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "switchpoint"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "<pymc.distributions.DiscreteUniform 'switchpoint' at 0x102452850>"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "`DiscreteUniform` is a subclass of `Stochastic` that represents\n",
      "uniformly-distributed discrete variables. Use of this distribution\n",
      "suggests that we have no preference *a priori* regarding the location of\n",
      "the switchpoint; all values are equally likely. \n",
      "\n",
      "Now we create the\n",
      "exponentially-distributed variables `early_mean` and `late_mean` for the\n",
      "early and late Poisson rates, respectively:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "early_mean = Exponential('early_mean', beta=1., value=3)\n",
      "late_mean = Exponential('late_mean', beta=1., value=1)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Next, we define the variable `rate`, which selects the early rate\n",
      "`early_mean` for times before `switchpoint` and the late rate\n",
      "`late_mean` for times after `switchpoint`. We create `rate` using the\n",
      "`deterministic` decorator, which converts the ordinary Python function\n",
      "`rate` into a `Deterministic` object."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "@deterministic\n",
      "def rate(s=switchpoint, e=early_mean, l=late_mean):\n",
      "    # Create a vector of Poisson means\n",
      "    out = np.empty(len(disasters_array))\n",
      "    out[:s] = e\n",
      "    out[s:] = l\n",
      "    return out"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The last step is to define the number of disasters `disasters`. This is\n",
      "a stochastic variable but unlike `switchpoint`, `early_mean` and\n",
      "`late_mean` we have observed its value. To express this, we set the\n",
      "argument `observed` to `True` (it is set to `False` by default). This\n",
      "tells PyMC that this object's value should not be changed:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "disasters = Poisson('disasters', mu=rate, value=disasters_array, \n",
      "                    observed=True)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Why are data and unknown variables represented by the same object?\n",
      "\n",
      "Since its represented by a `Stochastic` object, `disasters` is defined\n",
      "by its dependence on its parent `rate` even though its value is fixed.\n",
      "This isn't just a quirk of PyMC's syntax; Bayesian hierarchical notation\n",
      "itself makes no distinction between random variables and data. The\n",
      "reason is simple: to use Bayes' theorem to compute the posterior, we require the\n",
      "likelihood. Even though `disasters`'s value is known\n",
      "and fixed, we need to formally assign it a probability distribution as\n",
      "if it were a random variable. \n",
      "\n",
      "Remember, the likelihood and the probability function are essentially the same, except that the former is regarded as a *function of the parameters* and the latter as a *function of\n",
      "the data*."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Parents and children\n",
      "\n",
      "We have above created a PyMC probability model, which is simply a linked\n",
      "collection of variables. To see the nature of the links, we can examine any node's `parents` attribute:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "switchpoint.parents"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 11,
       "text": [
        "{'lower': 0, 'upper': 110}"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The `parents` dictionary shows us the distributional parameters of\n",
      "`switchpoint`, which are constants. Now let's examine the parents of `disasters` and `rate`:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "disasters.parents"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 12,
       "text": [
        "{'mu': <pymc.PyMCObjects.Deterministic 'rate' at 0x107dd3cd0>}"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rate.parents"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 13,
       "text": [
        "{'e': <pymc.distributions.Exponential 'early_mean' at 0x107dd3810>,\n",
        " 'l': <pymc.distributions.Exponential 'late_mean' at 0x107dd38d0>,\n",
        " 's': <pymc.distributions.DiscreteUniform 'switchpoint' at 0x102452850>}"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We are using `rate` as a distributional parameter of `disasters`\n",
      "(*i.e.* `rate` is `disasters`'s parent). `disasters` internally\n",
      "labels `rate` as `mu`, meaning `rate` plays the role of the rate\n",
      "parameter in `disasters`'s Poisson distribution. Now examine `rate`'s\n",
      "`children` attribute:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rate.children"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 14,
       "text": [
        "{<pymc.distributions.Poisson 'disasters' at 0x107dd3690>}"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Because `disasters` considers `rate` its parent, `rate` considers `disasters` its child. Unlike `parents`, `children` is a set (an unordered collection of objects); variables do not associate their children with any particular distributional role, so an index is not required. Try examining the `parents` and `children` attributes of the other parameters in the model.\n",
      "\n",
      "The following **directed acyclic graph** is a visualization of the parent-child relationships in the model. Unobserved stochastic variables `switchpoint`, `early_mean` and `late_mean` are open ellipses, observed stochastic variable `disasters` is a filled ellipse and deterministic variable `rate` is a triangle. Arrows point from parent to child and display the label that the child assigns to the parent."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pymc import graph, MCMC\n",
      "graph.dag(MCMC([switchpoint, rate, early_mean, late_mean, disasters]))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 15,
       "text": [
        "<pydot.Dot at 0x107dd3b90>"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "!dot MCMC.dot -Tpng -o images/dag.png\n",
      "from IPython.core.display import Image\n",
      "Image('images/dag.png')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAckAAAErCAYAAAC1s3j4AABAAElEQVR4Ae2dB9gURdLHC1EMBBUV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\n3a/o6JQAAAAASUVORK5CYII=\n",
       "prompt_number": 16,
       "text": [
        "<IPython.core.display.Image object>"
       ]
      }
     ],
     "prompt_number": 16
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "As the examples above have shown, PyMC objects need to have a name assigned, such as `switchpoint`, `early_mean` or `late_mean`. These names are used for storage and post-processing:\n",
      "\n",
      "-   as keys in on-disk databases,\n",
      "-   as node labels in model graphs,\n",
      "-   as axis labels in plots of traces,\n",
      "-   as table labels in summary statistics.\n",
      "\n",
      "A model instantiated with variables having identical names raises an\n",
      "error to avoid name conflicts in the database storing the traces. In\n",
      "general however, PyMC uses references to the objects themselves, not\n",
      "their names, to identify variables."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Variables' values and log-probabilities\n",
      "\n",
      "All PyMC variables have an attribute called `value` that stores the\n",
      "current value of that variable. Try examining `disasters`'s value, and\n",
      "you'll see the dataset we provided for it:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "disasters.value"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 17,
       "text": [
        "array([4, 5, 4, 0, 1, 4, 3, 4, 0, 6, 3, 3, 4, 0, 2, 6, 3, 3, 5, 4, 5, 3, 1,\n",
        "       4, 4, 1, 5, 5, 3, 4, 2, 5, 2, 2, 3, 4, 2, 1, 3, 2, 2, 1, 1, 1, 1, 3,\n",
        "       0, 0, 1, 0, 1, 1, 0, 0, 3, 1, 0, 3, 2, 2, 0, 1, 1, 1, 0, 1, 0, 1, 0,\n",
        "       0, 0, 2, 1, 0, 0, 0, 1, 1, 0, 2, 3, 3, 1, 1, 2, 1, 1, 1, 1, 2, 4, 2,\n",
        "       0, 0, 1, 4, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1])"
       ]
      }
     ],
     "prompt_number": 17
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "If you check the values of `early_mean`, `switchpoint` and `late_mean`,\n",
      "you'll see random initial values generated by PyMC:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "switchpoint.value"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 18,
       "text": [
        "array(60)"
       ]
      }
     ],
     "prompt_number": 18
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "early_mean.value"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 19,
       "text": [
        "array(3.0)"
       ]
      }
     ],
     "prompt_number": 19
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "late_mean.value"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 20,
       "text": [
        "array(1.0)"
       ]
      }
     ],
     "prompt_number": 20
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Of course, since these are `Stochastic` elements, your values will be\n",
      "different than these. If you check `rate`'s value, you'll see an array\n",
      "whose first `switchpoint` elements are `early_mean`,\n",
      "and whose remaining elements are `late_mean`:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rate.value"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 21,
       "text": [
        "array([ 3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,\n",
        "        3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,\n",
        "        3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,\n",
        "        3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,\n",
        "        3.,  3.,  3.,  3.,  3.,  3.,  3.,  3.,  1.,  1.,  1.,  1.,  1.,\n",
        "        1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,\n",
        "        1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,\n",
        "        1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,  1.,\n",
        "        1.,  1.,  1.,  1.,  1.,  1.,  1.])"
       ]
      }
     ],
     "prompt_number": 21
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "To compute its value, `rate` calls the function we used to create it,\n",
      "passing in the values of its parents.\n",
      "\n",
      "`Stochastic` objects can evaluate their probability mass or density\n",
      "functions at their current values given the values of their parents. The\n",
      "logarithm of a stochastic object's probability mass or density can be\n",
      "accessed via the `logp` attribute. For vector-valued variables like\n",
      "`disasters`, the `logp` attribute returns the sum of the logarithms of\n",
      "the joint probability or density of all elements of the value. \n",
      "\n",
      "Try examining `switchpoint`'s and `disasters`'s log-probabilities and\n",
      "`early_mean` 's and `late_mean`'s log-densities:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "switchpoint.logp"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 22,
       "text": [
        "-4.709530201312334"
       ]
      }
     ],
     "prompt_number": 22
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "late_mean.value = 5"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 23
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "disasters.logp"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 24,
       "text": [
        "-318.0083429843733"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "early_mean.logp"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 25,
       "text": [
        "-3.0"
       ]
      }
     ],
     "prompt_number": 25
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "late_mean.logp"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 26,
       "text": [
        "-5.0"
       ]
      }
     ],
     "prompt_number": 26
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "`Stochastic` objects need to call an internal function to compute their\n",
      "`logp` attributes, as `rate` needed to call an internal function to\n",
      "compute its value. Just as we created `rate` by decorating a function\n",
      "that computes its value, it's possible to create custom `Stochastic`\n",
      "objects by decorating functions that compute their log-probabilities or\n",
      "densities (see chapter :ref:\\`chap\\_modelbuilding\\`). Users are thus not\n",
      "limited to the set of of statistical distributions provided by PyMC."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Using Variables as parents of other Variables\n",
      "\n",
      "Let's take a closer look at our definition of `rate`:\n",
      "\n",
      "    @deterministic(plot=False)\n",
      "    def rate(s=switchpoint, e=early_mean, l=late_mean):\n",
      "        ''' Concatenate Poisson means '''\n",
      "        out = empty(len(disasters_array))\n",
      "        out[:s] = e\n",
      "        out[s:] = l\n",
      "        return out\n",
      "\n",
      "The arguments `switchpoint`, `early_mean` and `late_mean` are\n",
      "`Stochastic` objects, not numbers. If that is so, why aren't errors\n",
      "raised when we attempt to slice array `out` using a `Stochastic` object?\n",
      "\n",
      "Whenever a variable is used as a parent for a child variable, PyMC\n",
      "replaces it with its `value` attribute when the child's value or\n",
      "log-probability is computed. When `rate`'s value is recomputed,\n",
      "`s.value` is passed to the function as argument `switchpoint`. To see\n",
      "the values of the parents of `rate` all together, look at\n",
      "`rate.parents.value`."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Fitting the model with MCMC\n",
      "\n",
      "PyMC provides several objects that fit probability models (linked collections of variables) like ours. The primary such object, `MCMC`, fits models with a Markov chain Monte Carlo algorithm:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pymc.examples import disaster_model\n",
      "M = MCMC(disaster_model)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 27
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "In this case `M` will expose variables `switchpoint`, `early_mean`,\n",
      "`late_mean` and `disasters` as attributes; that is, `M.switchpoint` will\n",
      "be the same object as `disaster_model.switchpoint`.\n",
      "\n",
      "To run the sampler, call the MCMC object's `sample()` (or `isample()`,\n",
      "for interactive sampling outside of the IPython notebook) method with arguments for the number of iterations, burn-in length, and thinning interval (if desired):"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M.sample(iter=10000, burn=1000)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------      31%                  ] 3147 of 10000 complete in 0.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------60%---               ] 6053 of 10000 complete in 1.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------91%--------------    ] 9139 of 10000 complete in 1.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------100%-----------------] 10000 of 10000 complete in 1.6 sec"
       ]
      }
     ],
     "prompt_number": 28
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Accessing the samples\n",
      "\n",
      "The output of the MCMC algorithm is a `trace`, the sequence of\n",
      "retained samples for each variable in the model. These traces can be\n",
      "accessed using the `trace(name, chain=-1)` method. For example:\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M.trace('switchpoint')[1000:]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 29,
       "text": [
        "array([39, 39, 39, ..., 41, 41, 41])"
       ]
      }
     ],
     "prompt_number": 29
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The trace slice `[start:stop:step]` works just like the NumPy array\n",
      "slice. By default, the returned trace array contains the samples from\n",
      "the last call to `sample`, that is, `chain=-1`, but the trace from\n",
      "previous sampling runs can be retrieved by specifying the correspondent\n",
      "chain index. To return the trace from all chains, simply use\n",
      "`chain=None`.\n",
      "\n",
      "A node's chain can be accessed directly using its `trace` attribute:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M.early_mean.trace()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 30,
       "text": [
        "array([ 2.60698391,  2.60698391,  2.60428494, ...,  2.79480505,\n",
        "        2.72259973,  2.69514232])"
       ]
      }
     ],
     "prompt_number": 30
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Sampling output\n",
      "\n",
      "You can examine the marginal posterior of any variable by plotting a\n",
      "histogram of its trace:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.hist(M.trace('late_mean')[:])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 31,
       "text": [
        "(array([   20.,   151.,   706.,  1645.,  2300.,  2042.,  1255.,   623.,\n",
        "          215.,    43.]),\n",
        " array([ 0.55752905,  0.63378676,  0.71004447,  0.78630219,  0.8625599 ,\n",
        "         0.93881762,  1.01507533,  1.09133304,  1.16759076,  1.24384847,\n",
        "         1.32010619]),\n",
        " <a list of 10 Patch objects>)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEACAYAAAC6d6FnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAEm9JREFUeJzt3X+wXGddx/H3BwLDj1ZCLRP6S1sxSMMg1EqDorL+mBoY\npRWUAoooHQatUobR0ZZRe/9QRB2ZlnHacaDQohKnI1oLpIWWdoU60FhM20CINEocEtsgCEJVmES+\n/rEn7XK9uXdz7909N3ner5mdOfvsOef57s3mfPZ5zp7dVBWSpPY8qu8CJEn9MAAkqVEGgCQ1ygCQ\npEYZAJLUKANAkhq1aAAkOSPJHUk+leSTSS7t2ueS7Euyo7u9cGyby5Pcn2R3kvPH2s9NsrN77Krp\nPSVJ0iSy2HUASZ4KPLWq7klyAvAJ4ELgZcBXq+qt89bfBLwHeC5wGnAbsLGqKsl24FeranuSbcDb\nquqWqTwrSdKSFh0BVNWDVXVPt/wQ8GlGB3aALLDJBcDWqjpYVXuBPcDmJKcAJ1bV9m69dzMKEklS\nTyY+B5DkTOAc4ONd0+uT3Jvk2iTru7ZTgX1jm+1jFBjz2/fzSJBIknowUQB00z9/BbyhGwlcA5wF\nPAd4APjjqVUoSZqKdUutkOQxwHuBP6+qGwGq6vNjj78DeF93dz9wxtjmpzN657+/Wx5v379AX34x\nkSQtQ1UtNC2/qKU+BRTgWmBXVV051n7K2Go/Bezslm8CXp7ksUnOAjYC26vqQeArSTZ3+3wVcOMR\nnsSaul1xxRW913Cs1GVN1tRCXWuxpuVaagTwfODngPuS7Oja3gS8IslzgAI+C7yuO3jvSnIDsAs4\nBFxSj1R3CXAd8HhgW/kJIEnq1aIBUFV3svAo4eZFtnkz8OYF2j8BPOtoC5QkTYdXAi9hMBj0XcKC\n1mJd1jQZa5rcWqxrLda0XIteCDZrSWot1SNJx4Ik1GqfBJYkHb8MAElqlAEgSY0yACSpUQaAJDXK\nAJCkRhkAktQoA0CSGmUASFKjDABJapQBIEmNWvIHYaS1YPQzErPl91LpeGcA6BgyywPy7ANHmjWn\ngCSpUQaAJDXKAJCkRhkAktQoA0CSGmUASFKjDABJapQBIEmNMgAkqVEGgCQ1ygCQpEYZAJLUKANA\nkhplAEhSowwASWqUASBJjTIAJKlRBoAkNcoAkKRGGQCS1CgDQJIatWgAJDkjyR1JPpXkk0ku7dpP\nSnJrks8k+VCS9WPbXJ7k/iS7k5w/1n5ukp3dY1dN7ylJqyPJzG/SLC01AjgIvLGqngk8D/iVJGcD\nlwG3VtXTgQ9390myCbgI2ARsAa7OI6/qa4CLq2ojsDHJllV/NtKqqhnfpNlaNACq6sGquqdbfgj4\nNHAa8GLg+m6164ELu+ULgK1VdbCq9gJ7gM1JTgFOrKrt3XrvHttGktSDic8BJDkTOAe4C9hQVQe6\nhw4AG7rlU4F9Y5vtYxQY89v3d+2SpJ6sm2SlJCcA7wXeUFVfHZ+rrKpKsmrj17m5uYeXB4MBg8Fg\ntXYtSceF4XDIcDhc8X5StfixO8ljgPcDN1fVlV3bbmBQVQ920zt3VNUzklwGUFVv6da7BbgC+Ndu\nnbO79lcAL6iqX5rXVy1Vj9o0etMxy9fGrPsb9enrX8uRhKo66k8RLPUpoADXArsOH/w7NwGv7pZf\nDdw41v7yJI9NchawEdheVQ8CX0myudvnq8a2kST1YNERQJIfAD4C3Mcjb4cuB7YDNwDfBuwFXlZV\nX+62eRPwGuAQoymjD3bt5wLXAY8HtlXVpQv05whAC3IEIB3ZckcAS04BzZIBoCMxAKQjm8oUkCTp\n+GUASFKjDABJapQBIEmNMgAkqVEGgCQ1ygCQpEYZAJLUKANAkhplAEhSowwASWqUASBJjTIAJKlR\nBoAkNcoAkKRGGQCS1CgDQJIaZQBIUqMMAElqlAEgSY0yACSpUQaAJDXKAJCkRhkAktQoA0CSGmUA\nSFKjDABJatS6vgvQsSlJ3yVIWiEDQCtQM+zLwJFWm1NAktQoA0CSGmUASFKjDABJapQBIEmNMgAk\nqVFLBkCSdyY5kGTnWNtckn1JdnS3F449dnmS+5PsTnL+WPu5SXZ2j121+k9FknQ0JhkBvAvYMq+t\ngLdW1Tnd7WaAJJuAi4BN3TZX55Erhq4BLq6qjcDGJPP3KUmaoSUDoKo+CnxpgYcWujLnAmBrVR2s\nqr3AHmBzklOAE6tqe7feu4ELl1eyJGk1rOQcwOuT3Jvk2iTru7ZTgX1j6+wDTlugfX/XLknqyXID\n4BrgLOA5wAPAH69aRZKkmVjWdwFV1ecPLyd5B/C+7u5+4IyxVU9n9M5/f7c83r5/oX3Pzc09vDwY\nDBgMBsspUZKOW8PhkOFwuOL9pGrpL/RKcibwvqp6Vnf/lKp6oFt+I/DcqnpldxL4PcB5jKZ4bgO+\ns6oqyV3ApcB24APA26rqlnn91CT1qH+jc/uz/jK447m/UZ++/rUcSaiqo/7GxCVHAEm2Ai8ATk7y\nOeAKYJDkOYz+h3wWeB1AVe1KcgOwCzgEXDJ2RL8EuA54PLBt/sFfkjRbE40AZsURwLHDEcB0+vT1\nr+VY7gjAK4ElqVEGgCQ1ygCQpEYZAJLUKANAkhplAEhSowwASWqUASBJjTIAJKlRBoAkNcoAkKRG\nGQCS1CgDQJIaZQBIUqMMAElqlAEgSY0yACSpUQaAJDVqyd8EljQ7o5/anB1/grJtBoC0psz6d4/V\nMqeAJKlRBoAkNcoAkKRGGQCS1CgDQJIaZQBIUqMMAElqlAEgSY0yACSpUQaAJDXKAJCkRhkAktQo\nA0CSGmUASFKjDABJapQBIEmNWjIAkrwzyYEkO8faTkpya5LPJPlQkvVjj12e5P4ku5OcP9Z+bpKd\n3WNXrf5TkSQdjUlGAO8Ctsxruwy4taqeDny4u0+STcBFwKZum6vzyG/cXQNcXFUbgY1J5u9TkjRD\nSwZAVX0U+NK85hcD13fL1wMXdssXAFur6mBV7QX2AJuTnAKcWFXbu/XePbaNJKkHyz0HsKGqDnTL\nB4AN3fKpwL6x9fYBpy3Qvr9rlyT1ZMUngauqmO0vWUuSVsG6ZW53IMlTq+rBbnrn8137fuCMsfVO\nZ/TOf3+3PN6+f6Edz83NPbw8GAwYDAbLLFGSjk/D4ZDhcLji/WT0Bn6JlZIzgfdV1bO6+38IfLGq\n/iDJZcD6qrqsOwn8HuA8RlM8twHfWVWV5C7gUmA78AHgbVV1y7x+apJ61L/Ruf1Z/lsd7/310Wfw\n/9vxIQlVlaXX/GZLjgCSbAVeAJyc5HPA7wBvAW5IcjGwF3gZQFXtSnIDsAs4BFwydkS/BLgOeDyw\nbf7BX5I0WxONAGbFEcCxwxHA8dCnI4DjxXJHAF4JLEmNMgAkqVEGgCQ1ygCQpEYZAJLUKANAkhpl\nAEhSowwASWqUASBJjTIAJKlRBoAkNcoAkKRGGQCS1CgDQJIaZQBIUqMMAElqlAEgSY0yACSpUQaA\nJDXKAJCkRhkAktQoA0CSGrWu7wK0ckn6LkHSMcgAOG7UjPszdKRjnVNAktQoA0CSGmUASFKjDABJ\napQBIEmN8lNAUsP6+Ahx1aw/saYjMQCkpvnx4ZY5BSRJjTIAJKlRBoAkNcoAkKRGGQCS1KgVBUCS\nvUnuS7Ijyfau7aQktyb5TJIPJVk/tv7lSe5PsjvJ+SstXpK0fCsdARQwqKpzquq8ru0y4Naqejrw\n4e4+STYBFwGbgC3A1UkcgUhST1bjADz/g70vBq7vlq8HLuyWLwC2VtXBqtoL7AHOQ5LUi9UYAdyW\n5O4kr+3aNlTVgW75ALChWz4V2De27T7gtBX2L0lappVeCfz8qnogyVOAW5PsHn+wqirJYpca/r/H\n5ubmHl4eDAYMBoMVlihJx5fhcMhwOFzxfrJa38uR5ArgIeC1jM4LPJjkFOCOqnpGkssAquot3fq3\nAFdU1V1j+yi/J+Tojb7PpY9L+mfZ5/HeXx999vMc/T+++pJQVUf9PRvLngJK8oQkJ3bLTwTOB3YC\nNwGv7lZ7NXBjt3wT8PIkj01yFrAR2L7c/iVJK7OSKaANwN903ya4DviLqvpQkruBG5JcDOwFXgZQ\nVbuS3ADsAg4Bl/h2X5L6s2pTQKvBKaDlcQroeOivjz6dAjpezHwKSJJ0bDMAJKlRBoAkNcoAkKRG\nGQCS1CgDQJIaZQBIUqMMAElqlAEgSY0yACSpUQaAJDXKAJCkRhkAktQoA0CSGmUASFKjDABJapQB\nIEmNMgAkqVEGgCQ1ygCQpEYZAJLUqHV9FyCpLUlm2l9VzbS/Y4kBIGnGZnlAnm3YHGucApKkRhkA\nktQoA0CSGuU5gCmY9UkuSVoOA2BqPNElaW1zCkiSGmUASFKjDABJapQBIEmNMgAkqVEGgCQ1ygCQ\npEbNNACSbEmyO8n9SX5zln1Lkr7ZzAIgyaOBPwG2AJuAVyQ5e1b9L9dwOOy7hCMY9l3AAoZ9F7CA\nYd8FLGDYdwELGPZdwBEMV7yHJDO9HUtmeSXwecCeqtoLkOQvgQuAT0+z0yuvvJKHHnpo2dvffvvt\n3HnnnatY0WoZAoOea5hviDVNYog1TWrIyuta7avy57rbQgyAIzkN+NzY/X3A5ml3Ojf3h3zlKz8D\nnLis7asOMRx+beL1k9uX1Y8kzdosA6CXn+V51KPghBN2kTxuWdt/7WsP8rjH3Tvx+l//+uf5+teX\n1ZUkzVRm9XNpSZ4HzFXVlu7+5cA3quoPxtbxt9skaRmq6qjnn2YZAOuAfwJ+FPg3YDvwiqqa6jkA\nSdLCZjYFVFWHkvwq8EHg0cC1HvwlqT8zGwFIktaWXq4EXuqCsCSDJP+ZZEd3+62+axqra0eSTyYZ\n9l1Tkl8f+xvtTHIoyfqeazo5yS1J7un+Tr8wzXqOoq4nJ/mbJPcmuSvJM6dczzuTHEiyc5F13tbV\ne2+Sc6ZZzyQ1JXlGko8l+VqSX5t2PUdR1892f6P7kvx9ku9eAzVd0NW0I8knkvxI3zWNrffc7ljw\nkiV3WlUzvTGa/tkDnAk8BrgHOHveOgPgpjVW03rgU8Dp3f2T+65p3vo/AdzWd02MPiD9+4f/RsAX\ngXVroK4/An67W/6uGfytfhA4B9h5hMdfBGzrljcDH59mPRPW9BTge4HfBX5t2vUcRV3fBzypW96y\nRv5WTxxbfhaja5x6ralb59HA7cD7gZcutc8+RgAPXxBWVQeBwxeEzTfLKyomqemVwHurah9AVX1h\nDdQ0v76ta6CmB4Bv6Za/BfhiVR1aA3WdDdwBUFX/BJyZ5CnTKqiqPgp8aZFVXgxc3617F7A+yYZp\n1TNJTVX171V1N3BwmnUs0O9SdX2sqv6zu3sXcPoaqOm/xu6eAEz7eDDJawrg9cBfAf8+yT77CICF\nLgg7bd46BXx/N8TalmTTGqhpI3BSkjuS3J3kVWugJgCSPAH4ceC9a6CmtwPPTPJvwL3AG6Zc06R1\n3Qu8BCDJecC3M4MDySIWqrnPeo4VFwPb+i4CIMmFST4N3AxcugbqOY3RG59ruqYlT/D28aPwk5x1\n/kfgjKr67yQvBG4Ent5zTY8BvofRx1ifAHwsycer6v4eazrsJ4E7q+rLU6rlsElqehNwT1UNkjwN\nuDXJs6vqqz3X9RbgqiQ7gJ3ADuB/p1jTJOaPcv1ExiKS/DDwGuD5fdcCUFU3Ajcm+UHgzxhNLfbp\nSuCyqqqMvpRoyVmUPgJgP3DG2P0zGL37edj4waKqbk5ydZKTquo/+qqJ0bu1L1TV/wD/k+QjwLOB\naQXAJDUd9nKmP/0Dk9X0/cDvAVTVPyf5LKP/GHf3WVf3mnrN4ftdXf8yxZqWMr/m07s2LaA78ft2\nYEtVLTUNMlNV9dEk65J8a1V9scdSzgX+svtCupOBFyY5WFU3HWmDPqaA7gY2JjkzyWOBi4BvKjDJ\nhi7BDg/XM8WD/0Q1AX8L/ECSR3dTLpuBXT3XRJInAT/U1Tdtk9S0G/ixrrYNjA7+0z7QTvKaelL3\nGEleC/xdVS3/WwJX7ibg57t6ngd8uaoO9FjPuDX1jWZJvg34a+DnqmpP3/UAJHna2DHqewB6PvhT\nVd9RVWdV1VmMzgP88mIHf+hhBFBHuCAsyeu6x/8U+Gngl5McAv6b0TvcXmuqqt1JbgHuA74BvL2q\nphYAE/6dAC4EPtiNTKZqwpreDLwryb2M3mD8xpTDe9K6NgHXZfR1I59kNJc8NUm2Ai8ATk7yOeAK\nRtOIh19P25K8KMke4L+AX5xmPZPUlOSpwD8wOnn/jSRvADZNOyiXqgv4HeDJwDXdMfdgVZ3Xc00v\nBX4+yUHgIaZ8jJqwpqPfZ/fRIUlSY/xJSElqlAEgSY0yACSpUQaAJDXKAJCkRhkAktQoA0CSGmUA\nSFKj/g+UAk3yeE4+ZgAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107a2f090>"
       ]
      }
     ],
     "prompt_number": 31
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "\n",
      "PyMC has its own plotting functionality, via the optional `matplotlib`\n",
      "module as noted in the installation notes. The `Matplot` module includes\n",
      "a `plot` function that takes the model (or a single parameter) as an\n",
      "argument:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pymc.Matplot import plot\n",
      "plot(M)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Plotting switchpoint\n",
        "Plotting"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " late_mean\n",
        "Plotting"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " early_mean\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAlQAAAFwCAYAAAB+TeWUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107a46510>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAFwCAYAAAB6uDWpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nB0ov5vGyDREsAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107d8d190>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAFwCAYAAAB6uDWpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XmYFNXV+PHvYRHRCC4oKqIIooIbggIBxUaFKLKoxIWI\nikowUZQYoon+Ehlj8BVfNcE1vBoEN3BLjBrcdYxGo6IoGsQdIyCLC8iqA3N+f1TXdE1NdXd1d/VS\nM+fzPPNMd3VV3VtdM12n7z33lqgqxhhjjDGmcM3KXQFjjDHGmMbCAitjjDHGmIhYYGWMMcYYExEL\nrIwxxhhjImKBlTHGGGNMRCywMsYYY4yJSKjASkSai8g8EXk04LWEiKxOvj5PRH4bfTWNMSZFRDqK\nyPMi8h8ReVdELkwu315EnhaRD0TkKRHZ1rPNpSLyoYgsFJHBnuW9ROSd5GtTy3E8xpjGI2yL1QRg\nAZBu0qsXVPXg5M8foqmaMcakVQNcpKr7AX2B80WkG/Ab4GlV3Rt4NvkcEekOnAJ0B44BbhERSe7r\nVuAcVe0KdBWRY0p7KMaYxiRrYCUiuwFDgNsBSbdalJUyxphMVHWZqr6VfLwWeA/oAAwHZiZXmwkc\nn3w8ApilqjWqugj4COgjIrsA26jqa8n17vRsY4wxOQvTYvVH4GKgNs3rCvQTkbdFZE7ym6ExxpSE\niHQCDgZeBdqr6vLkS8uB9snHuwKLPZstxgnE/MuXJJcbY0xeMgZWIjIUWKGq80jfKvUm0FFVDwJu\nBB6OtorGGBNMRH4APARMUNU13tfUuV+X3bPLGFNSLbK83g8YLiJDgC2BNiJyp6qe4a7g/TBT1cdF\n5BYR2V5Vv/buSETsA86YJkhVi5IqICItcYKqu1TV/UK3XER2VtVlyW6+FcnlS4COns13w2mpWpJ8\n7F2+JKAs+/wypgnK5/MrY4uVql6mqh1VdU/gVOA5b1AFICLt3SRQEekNiD+o8uyvZD+TJk1q1OU1\nhWO08uJdnmrxYpHkZ85fgAWq+ifPS48AZyYfn0mqBf0R4FQR2UJE9gS6Aq+p6jLgWxHpk9zn6aRp\ndS/1e9eY/g5KcSxxP0+N5bw0luNQzf/zK1uLVYPPFgAROTf5BzwN+DHwcxHZBKzHCcCMMaaY+gOj\ngfkiMi+57FLgauB+ETkHWAScDKCqC0TkfpzRzZuA8zT1yXkeMANoDcxR1SdKdRCldMUVV5S7CqZA\n7jmcNGlSmWtiMgkdWKnqC8ALycfTPMtvBm6OvmrGGBNMVV8ifYv70Wm2uQq4KmD5G8AB0dWuMrkX\n46qqqvJWxOTNAqp4aLQzrycSiUZdXjnKtPKsPBN/jenvwI6l8jSW4yiEFNKPmFNBIlqqsowxlUFE\n0CIlr5eSfX5VPne+VztPJir5fn412hYrY4wxKVdccYXlWcWcncN4sBYrY0zRWIuVKRVrsTJRsxYr\nY4wxxpgys8DKGGOMMSYiFlgZY0wTYPk58WfnMB4sx8oYUzSWY2VKxXKsTNQsx8oYY4wxpswssDLG\nGGOMiYgFVsYY0wRYfk782TmMB8uxMsYUjeVYmVKxHCsTtSabY1VVVcU//vGPclejgbfffpvXX3+9\n3NUwxhhjTAnFOrCqra2t+5ZSaebNm8drr73WYLl9mzLGGGMar4oJrDZu3Mjo0aM56qijGDFiBGvW\nrKG2tpZBgwaRSCQYPHgwa9asAaB79+6cffbZTJw4sW77qVOnMmvWLAA++OADRo8eHVhOz549GT9+\nPL169WL69OmceeaZ9OjRg4ceegiAuXPncuSRRzJgwACuu+46AJ5++mkSiQS9e/dmypQpAMyYMYOR\nI0cyfPhwevfuzbJly+qV8+c//5mpU6dyzDHH8NlnnzFgwABOPfVUpkyZwt13383AgQPp1asXd999\nNwArV65k6NChJBIJTj/9dACeeOIJBgwYQP/+/Zk9e3ZUb7UxpkJVV1czbNiwouy7WPk5mer805/+\nlPfeey/ttjNnzuSLL76IvE6NleVYxYSqluTHKSq9G2+8UadPn66qqrNnz9Zrr71WVVXXr1+vqqp/\n/OMf9bbbblNV1TZt2uiqVatUVbWqqkofe+wxXbFihY4YMUJVVS+//HJ98sknA8vp3LmzLl68WNeu\nXatt2rTRlStX6qpVqzSRSKiq6lFHHVW372HDhuny5cvr6rB582Y99NBDdcOGDXrHHXfoOeeco6qq\nt956q95www31ypkxY4befPPNqqr66aefapcuXbSmpqbeMa1fv1579uypqqoXXXSR/vWvf63bvra2\nVvv37681NTW6adMm7d+/v27evDnje2hMpUn+35fsc6ZYP9k+v6JQU1Ojzz//vA4dOrToZUWpkDon\nEgmdO3duJPUAtBTnyTQd+X5+VUyL1YIFC7j11lsZOHAgN954I1999RXr1q1j7NixJBIJpk+fXvfN\nZq+99qJt27Z124oIO+64I82bN2fFihU899xzDBo0KLCc7bbbjg4dOrD11lvTtWtX2rVrR9u2bdm4\ncSMA8+fP5/jjj2fgwIF8/vnnLF68mLlz5zJo0CCOPPJIPvvsM1asWIGI0KNHDwA6duzIN99806As\n9XT7HXTQQbRo0QJwWqIGDhzIsccey8cffwzAwoULOeKII+rWX7lyJR988AGDBg3i6KOPZvXq1Xz5\n5ZeFvMXGmCL7+OOPOfbYYznkkEMYMGAA77//PgCPPvooffv2pWfPngwaNIgVK1YATo7o6aefzmGH\nHcYZZ5xRLwF77733rvufr62tpWvXrnz11VeB5Y4ZM4bzzjuPH/7wh3Tp0oXq6mrOPPNMunfvzlln\nnVW33lNPPUW/fv3o1asXJ598MuvWrQPgyiuvpHfv3hxwwAGce+65desnEgl+85vf0KdPH/bZZx9e\neumlBmWLCGvXruWkk06iW7du9XoLEokEb775JrW1tYwZM4YDDjiAAw88kD/96U889NBDzJ07l9NO\nO42ePXvWfQYbE3ctyl0BV7du3ejXr1/dP+WmTZt45JFH6Ny5M/fccw/XX399XVdgs2b140E3gDnt\ntNOYMGECvXv3Tpt75V0etE6PHj148MEHadOmDbW1tTRr1ozhw4czbdo0OnXqRK9everK834IeoMo\ngJYtW7J58+a65946T548mRdffBFVpUuXLnXH/8ILL3DCCSegqrRr1459992Xp556ipYtW7Jp06a6\nwMwYU5nGjRvHtGnT2GuvvXj11Vc577zzePbZZzn88MP597//DcDtt9/ONddcw7XXXgs4X6peeukl\nWrVqRXV1NeB8towePZp77rmHCRMm8Mwzz9CjRw922GGHwHJFhFWrVvHKK6/wyCOPMHz4cF555RW6\nd+/OoYceyttvv02HDh2YPHkyzz77LK1bt2bKlClcf/31/O53v2P8+PH87ne/A+CMM87gscceY+jQ\noYgImzdv5tVXX+Xxxx/niiuu4Omnn65Xtqoyb948FixYwC677EL//v15+eWX6devX91n5Lx581i6\ndCnvvPMOAN9++y1t2rThpptu4rrrrqNnz56RnwtjyqVirtTjxo1j3Lhx3HHHHQBMnDiRvn37ctVV\nVzFv3jzat2/PHnvsEbit+887bNgwxo0b1+AfP2jddI+vvvpqTjzxRGpra2nVqhV/+9vfGDlyJMcf\nfzwHHHAAbdq0abCNiDQI0n74wx9yxhln8NprrzF58uR6r5144okcdthh9OzZk+233x6ASy+9lDFj\nxjB16lQ6duzIXXfdxW9/+1sGDRpEs2bN2HHHHbnvvvsyv4nGmLJZu3Ytr7zyCieddFLdsu+//x6A\nzz//nJNPPplly5bx/fff07lzZ8D57Bg+fDitWrVqsL+zzz6bESNGMGHCBKZPn16v5SmIm+e0//77\ns/POO7PffvsBsN9++7Fo0SKmTJnC3Llz6devX13d3MfPPfcc//u//8v69ev5+uuv2X///Rk6dCjg\nfF6Bk5+6aNGiwLJ79+7NrrvuCjhfThctWlS3b4AuXbrwySefcOGFF3LccccxePDgutf8X0pNem5+\n1aRJk8pcE5NRPv2H+fxQgr7vjRs36oABA4pejuvee1Vra0tWXFFt3uz8GBMlmlCO1erVq3WXXXYJ\nfO2II47QRx99VFVVq6ur63I6q6qq6vJJVRvmKx177LH67LPPaufOnbU2w4fNmDFj9MEHH1RVJ6dz\n//33b/Dao48+qqNGjWqw7YYNG7R9+/a6ePHiujpdccUVqurkQL3xxhuqqrpy5Urt1KlTg+39dR4/\nfrzOnDmzwfZr167Vhx56SI8//ng9++yzG7xeKCzHykQs38+vismxKtSHH37I4MGDmTBhQt2yIUOG\nMHDgwLqfOXPmRFrmT34C334b6S4Dff+981NMvXrByJHFLcOYxqxNmzbsueeePPjgg4DzpXX+/PmA\n0/XltujMmDGjbhvN0lozduxYRo8ezcknn1zQ1DIiQt++ffnXv/5Vl9e5bt06Pvzww7rcph122IG1\na9fywAMP5F1OEFXlq6++YvPmzZx44olceeWVzJs3D4BtttmGb0vxIWpMCVVMV2ChunbtygsvvFBv\nWdSBVLn88IfQqhW8/HLxynjrLVi8uHj7N6YpuOeee/j5z3/OH/7wB2pqahg1ahQHHnggVVVVnHTS\nSWy33XZ1g2CgYRqB//mwYcM466yzsnYDutsGPXa1a9eOGTNmMGrUKL777jvAyffs2rUrP/3pT+u6\nEPv06ROqjHR1Dnp9yZIlnHXWWdTW1gJOygU4Sfc/+9nP2GqrrXj55ZfZcsstsx6nMZUu1C1tRKQ5\nMBdYrKoNJiwRkRuAY4H1wBhVnRewjoYpK05EYNUq8AxQLFo5LVpATU1xy2jXDlauLF4ZpumxW9oU\nZu7cuUycOLHBl8Z8NPb8nKZwS5vGfg4rTb6fX2EDq18CvYBtVHW477UhwHhVHSIifYCpqto3YB8W\nWBVQjgVWJo4ssMrf1VdfzZ///GfuvffeeongJlhTCKxMaRUtsBKR3YAZwGTgl/4WKxH5M/C8qt6X\nfL4QOEJVl/vWs8CqgHJatixunlXYwOqbb2D5cth33+LVxTQeFlhF66qrrmqQA3XyySdz6aWXlqlG\nlcMCKxO1YgZWDwBXAW2AXwUEVo8C/6OqLyefPwP8WlXf8K1XER9MUWqKgdUJJ8DDD0MjO5WmSCyw\nMqVigZWJWr6fXxmT10VkKLBCVeeJSCLTqr7ngX/ZVVVVdY8TiQSJRKZdVrZx48pdg/QmTIAddoDL\nL89tuzCDjtauza9Opmmorq6um+TSVJY45+d89NFHvPLKK6HWveuuuxos+9GPfsROO+0UdbVKLs7n\nsCnJ2GIlIlcBpwObgC1xWq0eUtUzPOv8GahW1dnJ502iK9ANQiqxxUoEtt3W6bbLpYwdd4TknTbS\nOvpoePZZa7Ey4ViLlYnC9OnTOf/8ybRokT7XbO1a54b2P/jB6HrLa2qe4NlnH6Z///5FraNpfIrS\nYqWqlwGXJQs4Aqcr8Azfao8A44HZItIXWOUPqkxDmzfDunXgmcg9o1ynsGks14DvvnOCymaNZsY1\nY0w+mjc/grVrp2dYwwms1q6t32LVtq0FVKa0cr1cKYCInCsi5wKo6hzgExH5CJgGnJdPRb79tmE3\n08aN+QUImzbBL36RTy1KZ8qU4rd0FUOpA7bWreFXvyptmcYYY0y+QgdWqvqCO9WCqk5T1Wme18ar\n6l6qepCqvplPRbp1A3/KVevWMHVq7vu64Yb8tvN65RUI6KqPzPXX57Z+U2qxeucdJ6gG5ziS9201\nxhTgiiuuqMvRMfFk5zAeKmbm9aVLYc2ahss//DD3fS2PoCPyggvgjTfg9NML31eQoGP1uu8++O9/\n4eKLi1N+vkoRsB14IFxxRSr5Ps5BojGVwhKe48/OYTw0ysyVvfYqfB9hW4jyvehn2/+ll8IllzRc\nvno13HxzfmUWWqdS2rCh3DUonV//Gp56qrB9iBT3lkfGGGPCqfjAKp+L/R57OL9V4bnn8is3bLK0\nan43Ys62/3SvP/88jB+fe3kAZ50FyfuvBoqiZeibb5z7DkapkgK+YrjmGrjttsL3k7z9nDHGmDKq\n+MAqH+6F+IMP4Kij8ttHLqPQ2rZ1Ap5c5BosRJFjNWMG/OMfue0nVxdcAAcfHO0+m0JXYGMPHk35\nWX5O/Nk5jIeKybGC4AtoIRec5I3U85JrV2C2+Z9y3X+xgolM5YY5Zm+9VJ1pI1p4/orcpHOTu6VL\nYdgwJ7cvH00hADX5s/yc+LNzGA+NusWqFPsoVo5VsUQ5H9QDDzhzTHlFdXH3vj9NoTVHBN5+G97M\na0ytMcaYSlHxgVU+F1V3m0IuyLkGILkGFP66vf66M/9WVNasgfnzs5ebK+9xBo3YLEarSTlaYv7z\nH/jii9KVF8UxNoUA1BhjKl3FB1aF8F6sttwyeDTd6tXB23oDqxUrYMmS7GXkwn8R7N0bZs0Kv34Y\nTz4ZzX6y+eqr1OPG0h21//7ODadLKcqgNwrp/jdMPFl+TvzZOYyHisqxAucivWYNdOqU/z6CLlDf\nfQevvQbnn59atnGjc0+9bLld/fvDokVQU9NwPf+27do5s77/9rf51THdfqNSaFegt17uMbRrF019\n33kHOnfOXGYphb03YxREUu/nL3/p/B/MnJl5m7/8BfbbD/r2jb4+X37p3DuysQTKxvJzGgM7h/FQ\ncS1WJ5wAe+7p3NwY4MYb8+8iy9YCsHlzuP18+WX4Onz1lTNrezaFjgoUyT4SMaiMKFusgi667rKt\nt879vB14YPCM+bmOuAwjkYBPP828TtB7tW6d81NM06fDnXdmX2/s2OJNINuU5hEzxpgoVVxg9c03\nzu8DDkgty3Um9UzBw6pV4YKLQpLX3WWq6WdYjyKJ/KOPct+m2Hk47rGvXx/cwpfNl1+m32cY8+eH\nW/+FF7JPqCnitAh5A+VDDkm1EA0dCrffHr5u2crKR6Zj3bQp+wz/xhhjolW2wOree2H33esvW7sW\n3n3Xebx4ceFlBF2s3MAtKpkubE8+CW3aBL9WiukWcm2xUoXDDgtfdhRBmip8/XXh+3EddFD+k8L6\nicCCBVBdnVq2cKGzDJw5wWbNSj2PojwvVZgwIf8uyd/9Lv3fX6GmTi1tV6kpnOXnxJ+dw3goW2D1\n/PPw+efpX/cO48+32yxsgDB8eMNbioQd7p+pxWrp0vTbbb99uLplku19yfV927wZ/vWvzF2k//xn\n5n0E5WBlcv/9sMMODbcvJGgLO5fW6NGZX3dbFTP9Ha1b57RqZfLqq9mDR2+OlZs0rurcUNztFs9V\npln2s8n2/v/iF3Zz7LiZNGmS5ejEnJ3DeChbYJXtg3uLLUpTD4BHH3XmZHJNmBA+kTxTvk2mSUOz\n3dzZX2Yh006E5e3CzFeu2y5bln9ZUdUhHff9e+kluPXW1PLa2tQtaMJMQtu3L/zqV7mXH8X5MMYY\nU1plGxWYy0U/l3UXLUp/QQq7nxtugF12Cbdu0A2f3XIvvTT9dlHkWEWdL1XIhXzbbZ3RbIXO5xWF\nQmbcD/L4486PN9gfNy63fYQdKOFV6QFVpdfPGGPKoWJbrPK1554wb17wa6r5XQwqdeLFbMeS76jA\nfN6j1audLq9cuwKL8d7me8Gvra0flPmD37FjoyvLz9sV6N/3kiVw/PH1X3Prman8Qt7bMMdlgVW8\nWH5O/Nk5jIeKm8cqCuvXO78L+eAPuiht2uQkLI8Y0fC1qFtqin3RUnVmTt977/zLDjqGQt+HKI47\nlxaru+9O5VoNGuS0LLnJ6mGmzcjXf//rdCN365Z+Hfe9ePFF+Pvf67+Wa4tZru66q7j7N6VnuTnx\nZ+cwHhpFi9XmzfDYYw2XF3KRDko8f+mlhi0H6Xz7bXB9wsyBNHy4c+H1yqf1KdM2P/sZ7LNPw/oB\nrFyZvY5hywu7TT5dZQAPPwyTJ9dflum89+6dyo8CuOii1OPnn3emYYjC009nnidr4EDo3j31fO7c\nhu+fm38WdDyvvdZw2bx58NZbudc1aHRfrjcVN8YY46jYwMrfpXTEEc50DEFefx2GDUu/j3QJ0t4y\nwgRhbktYNk8+CW3bNlx+223wgx80XO5egN06PPpouHIK8X//13CZW75/GowgN98cPDN4vi1WQd1s\nmVRXOy2IkyY5s9x7R8B5gyVw5tNyy3n99eAgHPILxNNtM3gwnHde+u38wUzQfRf32CN9GUHLrr0W\nDj4Y/vpX5/X773eWjxiR+W+3VSv429/Sv26MMSa8ipsgNJ1//jP99AzpLm7uBJrphvDneiH95S9z\nW98vXQuGt3Ur3YzXpRgVmIvx4+H99xsuz7eVcO7c3NYfONAJBtzyvIMI/K19/iAm14A6jAcfDG5F\nykW68/Xwww2Xeevt7/ocORI++yz1/JFHGr4nfv6/zaC6zJwJzzwTXAe/mpr8WyFNcVh+TvzZOYyH\nRpFjle4DfsYM53dVVfZ9hJnsMCiQyFe6i2gucwNlC5yefhouvDC3bcJq1ar+lBRBwgQtbn2yzT22\neTM0b+7MdO52YYbNpcp0zO4NpN97L9y+/Nz6nnSS8/vbb2Gbbeq/Bs4tai65JPucV0GC5g5z9/3y\ny877ku71dM/9wryXY8ZAhw5w7rnO8//+F/r0Sc23Bam/tz32gKOPzr5PUzqWnxN/dg7jIWuLlYhs\nKSKvishbIrJARP4nYJ2EiKwWkXnJnyy3IC48ebu21un6CFNGULeXv4x0Iwm9+8kk37ykbNzJIYNu\nTVJdnfm2MY89Vrx7vrVIE5Ln2wIUdC9Er1/8wvn905/m3nLoH90XdD69+U658B/v//5v+nX/x/Of\nU+h9HNO9z7lOjpttfy4353DJErj8cuex2yom4pyfCRNS63/xBbzxRm51MMaYxiBrYKWqG4GBqtoD\nOBAYKCKHBaz6gqoenPz5Q7b95jOPlXebFStyvwHtv/5Vv8spyvmOvBeVXGXqmtpuu/Tb3XVX/YlN\ng2R7n++7L/U4KOFeJLhLJ11gENXoSP9+5s9PPU7XxeRPYgdnFvZFi3KrUy5yaRnyds8V2nKYrpxC\npprw8tfvxRcbbuMNrsPmgRljTGMXKsdKVd3U1y2A5kDQDTpyulTkcmHZeeegOuVSmuOjj+CUU1LP\nr7km3HZRzILutWlT+q7HXI+r0ODw1FPzKyPdQAIvVSdpeuutGy7/4ANnhOX114er5z//mRrx5g2s\nvO/XbwPaSf/wh9SUBsW40Gd7/73deC+9lHndbH9XheQsZTv2dK+PGePc7inodW8XpP8cm8pj+Tnx\nZ+cwHkLlWIlIM+BNoAtwq6r6bzurQD8ReRtYAvwqYJ1Aq1cHj6DLXJ/6cwxl6xYJsnRpKrk90z5U\no79x82mnpUZseb37LhxwQG778h+jf+6lbF1sUfO/j2++2XBE2p13OhfsXPYDMGqU89sNZsIci/fc\npZtZPluuWCb+wOoPf4Arr0yVc9xxDbdRzZ5MHuSYY5y8uTCi6gqcOdPpbg4adRvUHSwCc+bkVrYp\nDcvPiT87h/EQtsWqNtkVuBswQEQSvlXeBDqq6kHAjUDAOKb63IvittumKzN4m113dR57gyKXPwDK\nNIS8Q4f6F+YoWjPS7ePrr+t3uaXL57rhBlgQKhxN8QcXv/lNuDoVi7+8ww9vuE5QUOVvjclUb28+\n23/+E7pqaaezWLIk/D788mkx7NIlv7LCjsgL4q5/7rlOiyk4o2zdiUD9+/Ofj6AgNihpHlKDPKIc\n7GGMMXGR06hAVV0tIv8ADgGqPcvXeB4/LiK3iMj2qlqvy7DKMzxvyZIEkEhb1saNwcu/+ML57XZF\nvfRS6kM/16TmMCMBc7mApVv3wAOdi7cb9KS7IAGcc0761zZtapisHiYxu1C5zMQeVfJ6pv20bh1u\nPb906xbSnRq0rTeXKoh3aoN8866yHffbbzdcNmyYM6Bh8mRo185pXXMHdviPI9OgCFfwl6JqLrqo\nOnCfxpTL2rVrWb16dc7bNWvWjG3cYb7GhJQ1sBKRdsAmVV0lIq2BQcAVvnXaAytUVUWkNyD+oArq\nB1buKK98/fznzu/DD3eG4AN88klu+wiaH8gvlwu3O3Tfz98iEtTaFsaPf9zw1ibuxfidd5xuxGLc\nIiZfhZSdadt8L9jp9hkmwE5n4cKGy/761/Dbf/llfuVme29Hjmy4vn9i1EyDJm69NXsdvDelTklQ\n/wuT5YNUCjc3p6l1J23cuDXDh5+c83aqm9l11w4sWlQ5Ta9N9RzGTZgWq12Amck8q2bAXar6rIic\nC6Cq04AfAz8XkU3AeiBrSnSULSvuhdb9QvK73+W+jwULnAAt1xnAvXbaKdx6bldMroKGr4s4s3Yf\neKBzccwWWAW970FdYRs2pFoHw/r0UydPzJXLnFwubx5Uunv1uesEzfzu2rjRSZD3Hm+6ACpbvleu\n3DLnz8/ttkNRtlhlWl/VGQjgXXb55U6S/49/HFzHoPKsRSpeKuFivCZo3pgQNhQwb8x33z2V55bv\n8/33w/Mutxgq4Rya7LIGVqr6DtAzYPk0z+ObgZujrVpD6S48bndYrvlJfr/4BfToUX9ZJQ0Z93f7\ngfOeuAGD99YtuQjqQt1qq9z340/IznSvPD9vQAbO+96vX/1lbuuQe0HPlCR93XXOKEHvbWXS3afx\n9dfD1zMM762Ugm5h5BXm3pFB8m31dB18cOqWOa7/+7/gwCpMAFVJ/yemcrVp04aWLbcmx0HkAKie\nHn2FjCmCks68ftxxcPfdztxMxcgFCpMXksm6dXDoofWXFfuCkUud080d5S7fYgs46qj6r6vWT0Te\nvDn3yUxzybHy8l6Q3da06dOjLTOdoJF+hf59hOWte7ZpKQYMSD0u5ojNoG4//3uUrvxZs4KDz2Lc\nGsg0fjU1a8gnsDImLkp6r8A5c5xWpYcfLmyIe1wFXbj++Mdo9xlURosWcOSRzuMpU8J3WbpuuSW/\nungDq7PPdqZZyJScD3Djjc7vqVPTr5PvRTzfLthcee9bWInS3Tvz1VfT51YFtZBddVXqcbFm+DfR\nsTmQ4s8sXtDKAAAgAElEQVTOYTyU5V6BJ5wA7dunnj/0UDlqEU6xv4m7t6zJV7bAyq2/O7mme2sS\nr6A5tbx++Uu46KLsdfEfi/e9mzEjONE7nUxBUJhzEhQ8uFMLFFu+k2WGabF6//3UvRJz4b1BtPve\n+M/X6tVO12nYz227ZU28WH5O/Nk5jIeStlhBKpfGe+EMyusI4p9o0pXLBTtXUQZWxbjth0j9lqF0\ngVXQbYGKzX9spUx2DjrOfAY15CPf4wweZVdfrlOKuH7609TjDh2c3+n+9pYvz68MY4wxZQisfvYz\n53c+AcXBBwcvz3Yz5krx3HPBywsJOETqH/+rrwavl8cULvXkc7H1H1euowzL5ZhjCts+32A50/xm\nhe7bK5eZ640xxuSm5IFV3ETZYhV0k2PV8F2h6ZLXP/449dwfQEV1n8OBA3Pf5vzz6z///PP8yvbL\npSswH08+mf+24Mxuno+gUZ9B/PNR5csCq6bF8nPiz85hPJQlxwriMwdOlIFVUMK+avgL+aJFDZct\nXZr55r6F3LjX6733nPmycpGu67ZQYUY1ljNoCDpPYYQJrGpr85sfLIgFVk2L5efEn53DeChbYBX2\n23ljEjSv05w5DedwykV1debX/aO8glrNwrrttvy3jVJUwWKlSXcPSa8nnyy8Rc1lgZUxxkSvbOFN\nmHySSlDsUYF5TkRcJ9vFMeiecflqrHMVee89WE5nnVXa8iywMsaY6JUtsGqKLVZBdtmluPvPNpVC\nLuIySMAraHoJP/+ksMVUaCAdJQusmhbLz4k/O4fxULauwLh8qDfWVprG7sornd8vv5x93VLm+x12\nWOnKymbjxnLXwJSS5efEn53DeChbu5EFVo5//7uw7ePyPpZLmGkiMiX/R23+/NKVlU26m1IbY4zJ\nX9kCq+22K1fJjYsFVsYYY0zlsEynLLbZptw1yMxaHYwxYVh+TvzZOYwH0RIlEYmIQqqsnXeGZctK\nUrQxpmwEVS1Ku6qITAeOA1ao6gHJZVXAWMCd7ewyVX08+dqlwNnAZuBCVX0qubwXMAPYEpijqhMC\nytJSfVY2ZiIC1ALF+JNw9xnVeXqfXXYZztKl70e0PxM3Ivl9fpWtxcqCKmNMge4A/DcgUuB6VT04\n+eMGVd2BU4DuyW1uEanrSL8VOEdVuwJdRaTAmxoZY5oy6wo0xsSSqr4IfBPwUtA3zBHALFWtUdVF\nwEdAHxHZBdhGVV9LrncncHwx6muMaRossDLGNDYXiMjbIvIXEdk2uWxXYLFnncVAh4DlS5LLGx3L\nz4k/O4fxULZ5rIwxpghuBX6ffHwlcB1wTvmqUzlsDqT4s3MYDxZYGWMaDVVd4T4WkduBR5NPlwAd\nPavuhtNStST52Lt8SdC+q6qq6h4nEgkSiUQUVTbGVIjq6mqqs92AN4SyjQo0xjQFxRsVCCAinYBH\nPaMCd1HVL5KPLwIOVdWfJJPX7wV643T1PQPspaoqIq8CFwKvAf8AblDVJ3zl2KjACNioQBMn+Y4K\ntBYrY0wsicgs4AignYh8DkwCEiLSA+fq+ilwLoCqLhCR+4EFwCbgPE+kdB7OdAutcaZbqBdUNRZu\nbo51J8WXncN4yNhiJSJbAi8ArYAtgL+r6qUB690AHAusB8ao6ryAdazFypgmp7gtVqViLVbRsBYr\nEydFabFS1Y0iMlBV14tIC+AlETlMVevuriYiQ3Ca1LuKSB+c5NG+uVbEGGOMMSbusk63oKrrkw+3\nAJoDX/tWGQ7MTK77KrCtiLSPspLGGGOMMXGQNbASkWYi8hawHHheVRf4VukAfO55vpj6o2yMMcaU\nmc2BFH92DuMha/K6qtYCPUSkLfCkiCRUtdq3mr8PMk0nd5XncSL5Y4xpPKqTP6bSWMJz/Nk5jIfQ\nowJVdbWI/AM4hPqfnEHzwwTOA1M/sDLGND4J6n9hsm/XxpimJWNXoIi0c28JISKtgUGAf8TfI8AZ\nyXX6AqtUdXkR6mqMMcYYU9GytVjtAswUkWY4QdhdqvqsiLhzw0xT1TkiMkREPgLWAWcVt8rGGGNy\nZXMgxZ+dw3iwmdcLMGECTJ1a7loYU8lsHiuTYvNYmTjJdx6rrKMCTXodOpS7BsYYY4ypJBZYFcC+\nwBpjjDHGqyIDq27dSlfWttvmv21tbXT1MMaYYrI5kOLPzmE8VGSO1VNPweDBRa5Q0sCB8Pzz+W17\n1VVw2WXR1sfvllvgvPOKW4YxxWM5VibFcqxMnFiOVZ42bcp/21J8ztpneWVqbzdtMsYYE6AiAysp\n4ffbQgKX7t2jq0c6FlhVpmL+jRbSPW2MMaa8KjKwKqQVKVeFjOw7/vjo6pFOnPO4/v73ctcgniyY\nNsVg+TnxZ+cwHkLf0qaU2rUrXVmlbB3LR5wvsu3awfjxcNNN5a5JNFq3hg0bnMeHHQYPPliccoLO\n+a9+BddeW5zyvNq1gy+/LH45pvRsUsn4s3MYD2Vpsdptt8yvH3KI8/vf/y5+XSrdlluWuwb5E6n8\nwDUXF16Yenz//cUrJ6iVslQB9gEHNFw2cmRpyjbGmMagLIFVmIvEHXdAz56VUZdMmjePph7pbLMN\nzJ1b3DIq0aGHlrsGDXmDxCgDRn9LVNDfZDlbLnfdtXxlG2NM3FRkjhXAmDHQsmW5a5FdsQMrVWgW\n8Vlq1Sra/RXDiy+GW6+Ux1LoeTjppODl/iCy0rp/45znZ1IsPyf+7BzGQ1lyrHK5cOy0E6xYUby6\nFGrLLeH774u3f9XoL2ylunAX0qoTNmAq5UXfPZ5837/OnYOXewO25s1h5kw4+WSYPRtOOSV7mR06\nwJIl+dUpjBZ5fkrsuCOsXBltXUz+LD8n/uwcxkPFdgW6Fi1KPS5lUntYI0bUf37kkdHuf+jQcO9X\nulysIUMKK79r18K29wZXQ4fmv590rT1bbBF+H0cdlX/5ALvvHvw32Ldv8PqjRtV/rgodOzZcr0uX\n1OPFi+HHP3YCxpNPDlevSy7J/HqhE8zmGyDHocXZGGOiVvGBlffbvP8DfrvtoqlPIcaOhXPOST1/\n9tlo9rvnns7vtm3DvV/pWhWCLorFaLEKE+CEaV0KCgL69AkOSADmz8++T9f++2fPF8rUOrP77vVb\nYNx9eZPavW68seGy//634bJs52PkyMzrZAt8dt458+vZ5NsqGHUXtjHGxEHFffS1bl3/eatW8Omn\nzmP//FbpWjFyUWgS8oABcPvthdfD76yzUo/DXNjSHUc5L27+UYFhArrLL4cvvqi/LNPo0HTda9Dw\nbynMKEVvoPbUU5nXdY/H3zLlLS9o/XT7SWfvves//+UvM5fjN2ZM5tezySUQP/jg1OM//amwck20\nLD8n/uwcxkPZ57Fy80NEnA/woItEp07O7x13hK+/Tq2z446Fl68KNTWV121xwgnw3nvO4zAXNu/7\n1ro1PPooHH104YFjuu2nT3cCvgsuSM3ttGRJ/QlXwwYWXi1bwvbbh6vboEEN6+otw9/6lCn4dINX\nb8tb27b11+vXr/7zbMcT5vhnzgwXOHu3zTVY3mqr3NYvZPs333Ru97NihU3TUGksPyf+7BzGQ8nb\nMw4/vP5Fwn/xyhQI+F/zPvfvJxf5JOe+/nr+5YWx//5w773OY/f9atkSevTIvq1IKp+o0K7AdAHn\nzjs7XaBbb50qZ5ddMu/r4ouzl+ev7w9+4PwOqrO/Wy+Xv6WgMtxt1q8P3r5Nm/rPo+hS7dOn/n7S\nna9M62Q6zpEjG7bcZRK0r0z3RQx6D554Inx5xhjT2JQ8sJo1K/jDOFsrxW23NcxZOf301OMw91fz\ndlP4vfFG9u29Cm0FyGTBguDlw4alv5inm2MpqCvOu49PPmm4r1mznN8TJ8LAgZnLC7oQp7uQh0ns\nD9vKFPSav65BAUi2YMttHc1WNmRvaQob2HlbXvNpYcy3VTIoPy3fYFEE/va3wupjjDGNQckDqw4d\ngj+83W6DdB/KY8c27PrZe+/UqLdMQRM4LS9z5qR/vVu34OXDhgUvz3QBmjo1/WtuC08m/rq4Ze23\nX8N7G/7P/zgz1XuDCu97mK3bKCggdYf49+qV20XSXXfwYOf3rrvm3h2YLbDq1Sv9tv45xcLWPZf3\nyyvoWBYuTF9+uvq0bg0//GHmcjZvTl9uLufoqqtSjw84IFywGybYat++NPfONPmz/Jz4s3MYDxUz\nKtBdluv90PbdN3if/gTffLkX69NOq788U2uFt2vJ77PP8q9LVRXcd1/9ZSLw8svw29/WX+Z97A9W\nrrkmeF1wbtXina/J+3r37g3rlO6irurcush93T/X1wUXBG+XrVvWP71FLsLeYidsvl3Q3/E++9Qv\nD1Ldsv4kdEiNMszUAgj1B264I0b95YSp55Ah8K9/wUsvOc/9QXy6rsh05yuIv8vUVIZJkyZZjk7M\n2TmMh7IEVkFD3t0Lx7nn5rYvb5Dgats282gxL/eik+7i1KeP883+7rvh//2/3PYZZJttwu0jSLNm\nTtC2//6pZSKZA4FmzZzkfG83oDfo9B73gAFOwrtr8+ZUMNWiBZx/fv5199fRP9v4gAGpcrwytfq4\n3XaTJweXmW+XVNicu7DdZjvskL4+bhDitvKl4w2s9toruI5B3dP+MmtrnST8/v0bvp7LnGCZdO4M\nX30Vzb6MMSZusgZWItJRRJ4Xkf+IyLsi0mDWHhFJiMhqEZmX/Plt0L7c1prq6lReiXtx8k+lEJbb\nouS9yKnC8OG57SfdRbhLl1QuypVXpkaG5ZuL4l4Qa2qcvLEwwsxhlK7+2YKLZs1S3Yu//339ucFq\na2HcOOfc1NQ4E1em2//YscH7HzQoeGJQ/4hOtxUmXSDlvgfeLjG3le6yy5zf/pYS/82+M7Wu5WP6\ndJgxI/3rLVs6Uw5cemn6ddwW12xfQr3/H+m6Atetq5936H89aFu/adMazhmW66hUCD+y0xhjGpsw\nLVY1wEWquh/QFzhfRIIykl5Q1YOTP38I2tHuuzu/t9vOuVVNvUJqcqh1AH9g5b84+Ifi+4Vp3RBJ\n5eDkcjG+5ZaG5Yg4wUiYlrWuXXNLqva+7n+fg9ZdvDj4tZ12cl4PCl79dbjpptRjb0Bz7LHO1A9e\nK1fCj35Uf1nLlsF/A/7Aat261Gv+nCp/y1VQ62Cm984/07lIcHK/a8QIOPNM5/Hvflf/tR13dHKn\nJkzIPJIzTE6XP8fK2w39zjv1jykomPSeN38Xdv/+cNBBqed77eWcM3/5XhddlL3OpvJYfk782TmM\nh6ydHqq6DFiWfLxWRN4DdgXe862aV8eLe1GYOBEOPDCfPRQmlyDFKyiwcuvvvuadHymd++5r2C3m\nt9NOqUlS09UvaCTg4sWpbqh0Mh2/f7BAWJlaaCD4tjBBuWDuci93SoegZO9sIzWzneug/DV/PlNY\nQYMhgsoPm+A+ZYrTHQ31g/H993dy7FxXXOH8L3nfY+99F/1/t6ee6vx4A7Bs9bz+evjjH8PV21QO\ny82JPzuH8ZBTjpWIdAIOBl71vaRAPxF5W0TmiEhAmnN9jz1W//khh2S/IHudcIKvAhr82OX94J82\nLfPrYeTSYjVwYKrry1/OIYc4v8NMF+EKCqIOOgj+/vf663XokP4egkH78qqpSZ9rtPvuqdyrvfZq\n2AXnb0kKI1tXpvt7/HjntzdJ3BVmwk53Pw88kHsdcxH27ylsi9Wuu6bysPzH7i2rZcuGwXSLFqn3\nJlsLpn9/rokTnRY676hCY4wxDYWeGlNEfgA8CExQ1bW+l98EOqrqehE5FngYCBgDldKpkzMv1fvv\np5+3KZ1PP2047UC2i6r3W3vQxSzdhfDrr4PLyVSe20rVvLnzeN994YUXgkfVufr1g9Gj4Sc/Sb9O\nELfezZql8srymSLBL1MCt3dk4xNP1M//6dUr/Kiw7bdPvb/ZAqvx4533MZcZ8v3nuV+/VKtUUL5Y\nlMIOUshlss8wc5hlEvbLgP99U3WC6YcfTr+NtVgZY4wjVGAlIi2Bh4C7VbXBx6uqrvE8flxEbhGR\n7VW1XlhSVVVV9ziRSDB+fKJBbkoY3nyjIO4F5O23U/kjd9yReZqHdLy5Ld71M23n5gE1bx4+dyyX\n+ZPSTQCai+XLnbmHctk+KKjxTy0xd274/e2wQ/jAqksX58e/3K9Xr9Rkr1de6czT5J6rESOcnKdi\nW7gw/9steb8AhJVIZJ/INRe5zj3m3aa6uprq6urcCzVF5+bmWHdSfNk5jIesgZWICPAXYIGqBt5W\nVUTaAytUVUWkNyD+oArqB1ZRC+oKdHOehgyBE0907mXn1Lfh9mFH1Z16qpPTkulic+yxzsXd2zXn\nnxjUv9+uXdPvL5NcjsXLHbWVy3D77bdPn+iej2y3ckm3/Be/aDjqc+ednRwrf7L7J584c5C5eUhR\ntOZlE9RNCdlb8vK9SfNee8FzzwW/li3HLkjY4+7bt+GyRCJBIpGoe26JtpXDLsbxZ+cwHsK0WPUH\nRgPzRWRectllwO4AqjoN+DHwcxHZBKwHTi1CXcvCf5G54AK48MLMF7t993USjb15U7vvnprbx7/t\nmjVOLtT99+dep3wv/v7cpTffDHcfQn8XbBQefTSVa+YXdHz+xGmAL74IXr9TJ2dai//8J7c63XRT\n/fnCCvXRRw1HgJ51VnT7D7LXXnDddeHX9w66CFruF8XfoTHGNDZhRgW+RJYkd1W9Gbg5nwpE8YH8\n3HNON9Eee7j1CV4vU1J1s2bOkPndd4ef/jR7mWG6R/zlpZvbJ9ONhrOVnW+LlcutY7ZbAhXDn/7k\n5GsFzXPlyvXvI2j97t0z57cFKWQy1CDebkzX9Onhtg3T/Rxkv/1yuwGzK2xXoAVTxhjTUOjk9UqW\nLr/E5X4D33lnmDcv+IbLIs4Emf77CRYyqWQ+o+PCcMu+887gexmGnZMLcsvtitpxx2VfJ4rAKtvr\n3bvD0qW5lZNruV7nnVd/brNi+Pjj3CbpfOCB1N9r2GPxrlfOvyMTjuXnxJ+dw3hoFIGVnz/J2vuh\n36NHcGDlivJ2JsUKrFxBs2znqtJbHaIOrLyuu86ZRuDZZ7PPN5bNBRc4LUSVIuwtnVzeUZLue7hx\nY/bpOsDJKcx0E2lTGexiHH92DuOh7IFV1Bf2a66BkSNTz2+7DXr3Dl73yCOdpHavo492brmzxx7O\n5JBhW6wOPjh1s13Xhg05VZ3Vq8OtF2a+pmyaNcv/Vi6lVMzAzx3ckG1i0TDatWs4c3s6cQlm3RGK\n6f5O3O5VfyuvMcY0ZWUPrKLWtWv9b+vp7mEHTkuFX7NmcMQRqefpLoJuPpfrzTcbrpNri1XU8xE1\nBlEHIZUQ1OQayLnne/Dg1MjWYgqTvL5uXXQ3bTbGmMak0WVG+O+VFiSXi2vQrUncmbCzeeml3OZ1\nCiuKFqu4KGaOVbnep0mT4K23wq/vnu+LL05NprvNNsULbPzvS1Di/1Zbhe82N5XB7jMXf3YO46Hs\nH41RXtxqa6PdX6EtQ0FBWRSaUmCVqzgc+9Zb17/xcT7efrvwvLB0vO9hU2odbewsPyf+7BzGQ9kD\nqyjF4aIahUwXu5tuym9SyEpVzOT1OOvYsXj7birvoTHGFEOj6wpsCjIFVuef78wO31jkepHPNgAg\nn6Bhxozct4kzC6yMMSZ/ZW+x8t7A14TTVLpnZs+GnXbKbZts99q7+GLnXn65OP308rYClvp825xU\njZPNgRR/dg7joeyBVdu25a5B5QjbUtBULnynnJL7NtkCsXHjct9ns2aZZ4dvbKzFqnGyi3H82TmM\nh7IHVuUIEoJuLxInc+bAqlXlrkVlmj0b1q4Nt+5BBzlTGFSyc86Bn/yktGVaYGWMMflrkoHVEUfE\nuzutU6dy16BytWnj/ITRrh08+WRx61Oo228vfZkWWBljTP7KHlgNHQqvvVbuWhhjXE2lq7mpsfyc\n+LNzGA9lD6y6doVZs8pdC2OMy1qsGie7GMefncN4sO+mxph6LLAyxpj8WWBljKnHugKNMSZ/9hFq\njKmnd28Lrhoju89c/Nk5jAfREg2PExEtVVlxNW0a/Oxn8R6xaIyXiKCqse9ctM+vaIgIUAsU40/C\n3WdU5+l9dtllOEuXvh/R/kzc5Pv5Zd9LjTHGGGMiYoGVMcYYY0xELLCqINtvX+4aGGMaK8vPiT87\nh/GQNcdKRDoCdwI74XRe/5+q3hCw3g3AscB6YIyqzvO9bjkKWajCkiWw227lrokx0bAcK+NlOVYm\nToqZY1UDXKSq+wF9gfNFpJuv8CHAXqraFRgH3JprRaJWXV0du/JEcguq4niMVl7TKc8YY5qirIGV\nqi5T1beSj9cC7wG7+lYbDsxMrvMqsK2ItI+4rjlpChetxn6MVl68yzPGmKYop1vaiEgn4GDgVd9L\nHYDPPc8XA7sBywuomzHGmIhEdZ+5hx/+O2+++WYUVTI5snsFxkPowEpEfgA8CExItlw1WMX33BIS\njDGhicidwCxVfbzcdWmMoroY33ffo8yevQLolcfWVZHUoamygCoeQk0QKiItgceAx1X1TwGv/xmo\nVtXZyecLgSNUdblnHQu0jGmCwiZ/ikgr4BTgOOBl4HZVXVfMuoVlyespo0aNZfbsvsDYclfFx5LX\nTbTyTV7P2mIlzjCOvwALgoKqpEeA8cBsEekLrPIGVRD+w9UY02TtAHQGVuOkEUzHCbSMMSY2wnQF\n9gdGA/NFxJ1C4TJgdwBVnaaqc0RkiIh8BKwDzipKbY0xjdlE4BZV/RhARD7Psr7JgeXnxJ+dw3jI\nGlip6kuEGz04PpIaGWOaqmpPUHWcqv6j3BVqTOxiHH92DuOhJDOvi8gxIrJQRD4UkV/nuY/pIrJc\nRN7xLNteRJ4WkQ9E5CkR2dbz2qXJ8haKyGDP8l4i8k7ytakZyusoIs+LyH9E5F0RubAEZW4pIq+K\nyFsiskBE/qfYZSbXbS4i80Tk0RIc4yIRmZ8s77USlLetiDwoIu8l39M+xSpPRPZJHpf7s1pELizy\n8V2a/Bt9R0TuFZFWJfh7mZBc910RmZBcFkWZR3iKOTxTHYwxpmKpalF/gObAR0AnoCXwFtAtj/0c\njjPVwzueZdcAlyQf/xq4Ovm4e7KclslyPyKVqP8a0Dv5eA5wTJrydgZ6JB//AHgf6FbMMpOvb5X8\n3QL4N3BYCcr8JXAP8EgJ3tdPge19y4pZ3kzgbM972rbY72dynWbAF0DHYpWX3OYToFXy+X3AmUV+\nP/cH3gG2xPnffhroElGZTwNHAUcCdxT7synHzx81jlNPPUfhNnXuFVFJPyR/otrfQt1ll73L/Xab\nMkr+3+f8eVGKFqvewEequkhVa4DZwIhcd6KqLwLf+BbXTUya/H188vEInGHbNaq6COfDvI+I7AJs\no6qvJde707ONv7ygiVE7FLPMZFnrkw+3wLlwfVPMMkVkN2AIcDupYTVFPUZPOa6ilCcibYHDVXU6\ngKpuUtXVJTg+gKNx/u4/L2J53+LcGWErEWkBbAUsLfLx7Qu8qqobVXUz8AIwMqIyPwf2TpbxizTl\nmzzZfebiz85hPOQ0QWiegiYP7RPRvttravThcsCd7X1XnNYeb5kdcC5Ciz3LlySXZyT1J0Ytapki\n0gx4E6cV4FZV/Y+IFLPMPwIXA208y4pZngLPiMhmYJqq3lbE8vYEVorIHcBBwBs4F+xS/N2cCsxK\nPi5Kear6tYhcB/wX2AA8qapPF/nv5V1gsohsD2zECcrnRnSMnYEPgFbABOD3aepg8mD5OfFn5zAe\nShFYlWTyF1VVKcJcWeJMjPoQzsSoa0RSjS3FKFNVa4EeydaWJ0VkoO/1yMoUkaHAClWdJyKJNPWJ\n+hj7q+oXIrIj8LQ4c54Vq7wWQE9gvKq+LiJ/An5TxPIAEJEtgGE4XWL1RHz+uuAEip1wpih4QERG\nF6u85P4WisgU4CmcEcBvAZsjKrMzznx5NQVX1BhjyqQUXYFLcPJMXB2p/021EMtFZGeAZNfCijRl\n7pYsc0nysXf5knQ7F2di1IeAu1T14VKU6Up2Wf0DZ3rjYpXZDxguIp/itK4cKSJ3FfMYVfWL5O+V\nwN9wuoqLVd5iYLGqvp58/iBOoLWsyOfwWOCN5DFSxOM7BHhZVb9S1U3AX4EfFvv4VHW6qh6iqkfg\ndFV/ENExLlbVd1X1fVW1WRmNMbFUisBqLtBVRDolv8mfgjOhaBQewUnWJfn7Yc/yU0VkCxHZE+gK\nvKaqy4BvxRkZJsDpnm3qSb4eNDFqMcts546mEpHWwCBgXrHKVNXLVLWjqu6J03X1nKqeXqzyRGQr\nEdkm+XhrYDBOInSxjm8Z8LmI7J1cdDTwH+DRYpTnMYpUN6C732KUtxDoKyKtk+sdDSwo9vGJyE7J\n37sDJwL3RnSMzUXkURF5QEQeSFe+yY/l58SfncOYyCfjPdcfnG/w7+Mkrl6a5z5m4STmfo+Ts3UW\nsD3wDM435qeAbT3rX5YsbyHwI8/yXjgX84+AGzKUdxhQi9PVMS/5c0yRyzwAJ7/qLWA+cHFyedHK\n9Kx/BKlRgUUpDyfn6a3kz7vu30KR39ODgNeBt3FadNoWubytgS9xErMpwfFdghMsvoOTNN6y2H8v\nwD+TZb4FDIzqGHFG3x6afG23EH+z03HyubwjhbfHGV0YVI9LgQ+T9RgcUI8PgalpylLjsFGBpqkg\nz1GBoe4VaIwxxSYitwHfq+r5InKLqp6XZf3DgbXAnap6QHLZNcCXqnqNOHPmbaeqvxGR7jgta4fi\nJM8/A3RVVRVnPrXxqvqaiMzBCSyf8JWl9lnpsHsFmqZC8rxXYEkmCDXGmBDW4rRAgTPKMSMt8RQs\nxhgThgVWxphK8SXQLzmFRG2e+8g07YN30Iw77YN/eagpWOLI8nPiz85hPJRiugVjjMlKVSeLyL5A\nM1VdEMH+Ip1qoqqqqu5xIpEgkUhEteuSsDmQ4s/OYXFVV1dTXV1d8H4ssDLGVAQRcUdStk7mNuTT\nJa9fckQAAB9XSURBVLdcRHZW1WVRTzXhDaxM07Bu3TfceOONeW3br18/evXqFXGNTDH5vzDl2zpo\ngZUxpiKo6iiom+rkojx34077MIWG0z7cKyLX43T1udM+qIh8KyJ9cO5deDrOCEXT5G3L+vWncPHF\nH+S8ZW3tv/jDHzZYYNVEWWBljKkIIrIfzpCulsB+IdafhTNNSDsR+Ry4HLgauF9EzgEWAScDqOoC\nEbkfZ56vTcB5nmF+5wEzgNbAHP+IwMbC/fZt3UlhtWfTphvZtCn3LVu0uCT66mDnMC4ssDLGVIof\nJ39/R4hWI7eFK8DRada/CrgqYPkbOHPINWp2MY4/O4fxYIGVMaZSzPU83k1EdlPVf5StNsYYkwcL\nrIwxlWIs8C+c7sDDyHzbIGOMqUgWWBljKsVCVb0WQER2VNWZ2TYw4Vl+TvzZOYwHC6yMMRVDRP6C\n02K1PNu6Jjd2MY4/O4fxYIGVMaZS/D+ceaRW4SSwG2NM7NgtbYwxleJPwCRV/RbIb1ZGY4wpMwus\njDGVohb4LPl4VTkr0hjZfebiz85hPFhXoDGmUnwHdBeRC4Dtyl2Zxsbyc+LPzmE8WGBljCm75G1s\nHgTaAQLcUt4aGWNMfiywMsaUXfKefQNV9Zpy18UYYwphgZUxpuxEZAQwQkR+BHwNoKonlbdWjYvN\ngRR/dg7joWSBlYho9rWMMY2NqkqI1Y5R1f4icquq/rzolWqC7GIcf3YO46GkowJVtVH8TJo0qex1\nsOOwY4nDTw52F5Hjkr+HiMiQIn0MGWNMUVlXoDGmEjyAk7h+P7BjmetijDF5s8DKGFN2qjqj3HVo\n7Cw/J/7sHMaDBVZ5SCQS5a5CJBrLcYAdizHZ2MU4/uwcxoPNvJ6HxnLhayzHAXYsxhhjKoMFVsYY\nY4wxEckaWInIdBFZLiLvZFjnBhH5UETeFpGDo62iMcaYQtl95uLPzmE8hMmxugPnTvN3Br2YHBa9\nl6p2FZE+wK1A3+iqaIwxplCWnxN/dg7jIWuLlaq+CHyTYZXhwMzkuq8C24pI+2iqZ4wxxhgTH1Hk\nWHUAPvc8XwzsFsF+jTHGGGNiJarkdf8tK+z2NcYYU0EsPyf+7BzGQxTzWC0BOnqe75Zc1kBVVVXd\n40QiYcPKjWlkqqurqa6uLnc1TADLz4k/O4fxEEVg9QgwHpgtIn2BVaq6PGhFb2BljGl8/F+Y7Nu1\nMaapyRpYicgs4AignYh8DkwCWgKo6jRVnZO8aepHwDrgrGJW2BhjjDGmUmUNrFR1VIh1xkdTHWOM\nMcVg95mLPzuH8RD7ewVWVVVx6KGHctxxx5W7KsYYU7HsYhx/dg7jIda3tKmtrUXEPyCxcVHVwMfG\nGGOMqTwVE1ht3LiR0aNHc9RRRzFixAjWrFlDbW0tgwYNIpFIMHjwYNasWQNA9+7dOfvss5k4cWLd\n9lOnTmXWrFkAfPDBB4wePTqwnJ49ezJ+/Hh69erF9OnTOfPMM+nRowcPPfQQAHPnzuXII49kwIAB\nXHfddQA8/fTTJBIJevfuzZQpUwCYMWMGI0eOZPjw4fTu3Ztly5bVK+fdd98lkUjQr18/LrjgAsAJ\njM4//3wGDBjAkUceyZdffsk777zD4YcfzmGHHcbVV18NOK1wY8aM4bjjjmP+/PkMGDCAU089ta5s\nY4wxxlQoVS3Jj1NUejfeeKNOnz5dVVVnz56t1157raqqrl+/XlVV//jHP+ptt92mqqpt2rTRVatW\nqapqVVWVPvbYY7pixQodMWKEqqpefvnl+uSTTwaW07lzZ128eLGuXbtW27RpoytXrtRVq1ZpIpFQ\nVdWjjjqqbt/Dhg3T5cuX19Vh8+bNeuihh+qGDRv0jjvu0HPOOUdVVW+99Va94YYb6pWzYcOGuscj\nRozQDz/8UP/+97/rBRdcULe8trZWhw0bpgsXLlRV1cGDB+uiRYu0qqpKL7/8clVV/fTTT7VLly5a\nU1OT8f0zphIl/+9L9jlTrJ9sn19xUFVVpVVVVQXv59RTz1G4TUEr7IfkT7nrodqixcU6ZcqUCM5a\nfVGdQxNOvp9fFZNjtWDBAubOncudd95JTU0NAwYMYN26dYwbN44lS5bw9ddfc9JJJwGw11570bZt\n27ptRYQdd9yR5s2bs2LFCp577rm0Uztst912dOjQAYCuXbvSrl07wGkxA5g/fz7HH388AKtWrWLx\n4sWsW7eO3//+99TU1PDZZ5+xYsUKRIQePXoA0LFjR95444165XzyySf86le/Yv369XzyyScsXbqU\nhQsXcsQRR9Sr97Jly9hnn30ApzXt448/BuCQQw6pW++ggw6iRYuKOVXGmBiy/Jz4s3MYDxVzte7W\nrRv9+vWr68LbtGkTjzzyCJ07d+aee+7h+uuvr+sKbNasfg+mE1jCaaedxoQJE+jdu3fa3Cvv8qB1\nevTowYMPPkibNm2ora2lWbNmDB8+nGnTptGpUyd69epVV567vRulev35z39m4sSJdV2bqkq3bt14\n5plnGDlyJODkiLVv356FCxeyzz778Oabb/Kzn/2MF198sd4x+o/XGGOMMZWpYgKrcePGMW7cOO64\n4w4AJk6cSN++fbnqqquYN28e7du3Z4899gjc1g1whg0bxrhx43j66afTlpMusHIfX3311Zx44onU\n1tbSqlUr/va3vzFy5EiOP/54DjjgANq0adNgGxFpEKQNGzaMCRMmsO+++6KqiAjDhg3jiSee4PDD\nD6dly5bcf//9TJ48mbFjx6KqDB06tO4YM+3bGGOMMZVJ/C0tRStIRItd1nfffcfgwYN54YUXilqO\nMSYcEUFVY//NoBSfX8UW1RxIo0aNZfbsvsDYCGoVJffPrPznqUWLS5g8uR2XXHJJpPu1eaxKK9/P\nr4ppsSrUhx9+yNixY5kwYULdsiFDhrBhw4a65xdffDFDhgwpR/WMMaas7GIcf3YO46HRBFZdu3Zt\n0FI1Z86cMtXGGGOMMU2RZUUbY4wxxkTEAitjjGkCrrjiirocHRNPdg7jodF0BRpjjEnP8nPiz85h\nPFiLlTHGGGNMRCywMsYYY4yJiAVWxhjTBFh+TvzZOYwHy7EyxpgmwPJz4s/OYTxkbbESkWNEZKGI\nfCgivw54vZ2IPCEib4nIuyIypig1NcYYY4ypcBkDKxFpDtwEHAN0B0aJSDffauOBearaA0gA14mI\ntYQZY4wxpsnJ1mLVG/hIVRepag0wGxjhW+cLwL0zcRvgK1XdFG01jTHGFMLyc+LPzmE8ZGtZ6gB8\n7nm+GOjjW+c24DkRWQpsA5wcXfWMMcZEwfJz4s/OYTxka7EKc5vwy4C3VHVXoAdws4hsU3DNjDHG\nGGNiJluL1RKgo+d5R5xWK69+wGQAVf1YRD4F9gHm+ndWVVVV9ziRSJBIJHKusDGmclVXV1NdXV3u\nahhjTNlkC6zmAl1FpBOwFDgFGOVbZyFwNPAvEWmPE1R9ErQzb2BljGl8/F+YLB+kcrjnwrqT4svO\nYTxkDKxUdZOIjAeeBJoDf1HV90Tk3OTr04CrgDtE5G2crsVLVPXrItfbGGNMDuxiXFqbN29m06b8\nxnE1b94cEWmw3M5hPIhqmDSqCAoS0VKVZYypDCKCqja8QsSMfX6ljBo1ltmz+wJjy10VH/fPrPzn\nSeQ3wLV5bau6mS+//JIddtgh2kqZnOX7+WW3tDHGGGMipHo1qpvy+mnVavtyV98UyAIrY4xpAmwO\npPizcxgPNkO6McY0AZafE392DuPBWqyMMcYYYyJigZUxxhhjTEQssDLGmCbA8nPiz85hPFiOlTHG\nNAGWnxN/dg7jwVqsjDHGGGMiYoGVMcYYY0xELLAyxpgmwPJz4s/OYTxYjpUxxjQBlp8Tf3YO46Es\nLVZVVVXlKNYYY4wxpqjKElhZU6YxxhhjGiPLsTLGmCbA8nPiz85hPFiOlTHGNAGWnxN/dg7jwVqs\njDHGGGMikjWwEpFjRGShiHwoIr9Os05CROaJyLsiUh15LY0xxhhjYiBjV6CINAduAo4GlgCvi8gj\nqvqeZ51tgZuBH6nqYhFpV8wKG2OMyZ2bm2PdSfFl5zAesuVY9QY+UtVFACIyGxgBvOdZ5yfAQ6q6\nGEBVvyxCPY0xxhTALsbxZ+cwHrJ1BXYAPvc8X5xc5tUV2F5EnheRuSJyepQVNMYYY4yJi2wtVhpi\nHy2BnsBRwFbAKyLyb1X9sNDKGWOMMcbESbbAagnQ0fO8I06rldfnwJequgHYICL/BA4CGgRW3hnX\nq6urSSQSudfYGFOxqqurqa6uLnc1EJFFwLfAZqBGVXuLyPbAfcAewCLgZFVdlVz/UuDs5PoXqupT\n5ah3MVl+TvzZOYwHUU3fKCUiLYD3cVqjlgKvAaN8yev74iS4/whoBbwKnKKqC3z7UrcsESFTucaY\nxiH5vy5lKPdToJeqfu1Zdg3Ol8BrkiOct1PV34hId+Be4FCcVIdngL1VtdazrdpnlmPUqLHMnt0X\nGFvuqvi4f2bxPk+tWu3AkiUfsMMOO5S7Kk1evp9fGXOsVHUTMB54ElgA3Keq74nIuSJybnKdhcAT\nwHycoOo2f1BljDFl4P9AHA7MTD6eCRyffDwCmKWqNcmBOh/hDNwxxpicZZ15XVUfBx73LZvme34t\ncG20VTPGmLwp8IyIbAamqeptQHtVXZ58fTnQPvl4V+Dfnm2DBukYY0wodksbY0xj1F9VvxCRHYGn\nRWSh90VVVRHJ1GcU7/6kAJafE392DuPBAitjTKOjql8kf68Ukb/hdO0tF5GdVXWZiOwCrEiu7h+k\ns1tyWT3ewTeJRCJ2g2/sYhx/dg6LK6rBNxmT16NkyevGND3lSF4Xka2A5qq6RkS2Bp4CrsC5g8RX\nqjpFRH7z/9u7/2C5yvqO4+8PCSRACNQyBQlRoNI2/MGv2hCgaqzipLGGOrViKrVVbCkMtAwjBRzG\nLO0MaEVBZaqI+KMRoUgpkAEBLaztGMOPQpCahBIhLZEhotDwI0MmV77945xNDsvu3bN798d57n5e\nM2funrPP2X2e+5xz9rvnfPc5wH5NyesL2ZW8/qZitrqT13dx8vpgOXm9Ono9fvmMlZlNNwcA/yoJ\nsmPctRFxl6QHgBsknUY+3AJARKyTdAPZD3QmgDMdRZlZrxxYmdm0EhFPAEe3WP4s2VmrVutcAlwy\n4KqNlPNz0uc+TIMDKzOzMeAP4/S5D9PQ6V6BZmZmZlaSz1iZmY2Z9evXc+aZ5zEx0f26GzY8DCzq\ne53MpgsHVmZmY6CYn/Pcc89x330b2bat13Gdj+xfxaw051ilwcMtmNnAjOpegf023YZbWL16NUuX\nfoytW1ePuip95OEWrL8Gcq9AMzMzMytvpIFVcSRjMzMzs9SNNLBqXC82M7PBuvjii33MTZz7MA0j\nzbFyrpXZ9OYcq2pyjlV1OceqOpxjZWZmZjZiDqzMzMzM+sSBlZnZGHB+Tvrch2nomGMlaQlwBTAD\n+EpEfKpNud8Bfgi8PyJuavG8c6zMxoxzrKrJOVbV5Ryr6hhIjpWkGcCVwBLgCGC5pAVtyn0KuINd\nW7eZmZnZWOl0KXAhsDEiNkXEDuB64OQW5c4GbgSe6XP9zMzMzJLRKbCaBzxZmN+cL9tJ0jyyYOuL\n+aK0z8OamU1Dzs9Jn/swDZ1uwlwmSLoCuCAiQpKY5FJgcaT1er1e4qXNLCX1et37dkX5xr3pcx+m\nYdLkdUmLgFpELMnnLwReKSawS3qcXcHU/sA24C8i4tam13LyutmYcfJ6NTl5vbqcvF4dvR6/Op2x\negA4XNIhwFPAKcDyYoGIOKxQia8Bq5qDKjMzM7NxMGlgFRETks4C7iQbbuGaiFgv6fT8+auGUEcz\nM5uiRm6OLyely32YBt8r0MwGxpcCq8mXAqvLlwKrw/cKNDMzMxsxB1ZmZmZmfVKZwKo4FIOZmfWX\nx0BKn/swDZXJsXK+ldn04xyranKOVXU5x6o6nGNlZmZmNmKdxrEyMzOzITrjjHOZNWt21+stXHgU\nZ5995gBqZN1wYGVmNgY8BlIatm//DN/+9vaWz9VqT+d/D2zx7Fo2b77LgVUFOLAyMxsDDqhS8edt\nn5n8N143A1/va02sN86xMjMzM+sTB1ZmZmZmfeLAysxsDHgMpPTVahdTq7kPq845VmZmY8A5Vumr\n1dyHKfAZKzMzM7M+cWBlZmZm1icOrMzMxoBzrNLnHKs0VC7Hqlar+YbMZmZ95hyr9DnHKg2lzlhJ\nWiJpg6THJJ3f4vkPSnpY0o8k/UDSkb1WyN+ozMzMLFUdAytJM4ArgSXAEcBySQuaij0OvDUijgT+\nHvhyvytqZmZmVnVlzlgtBDZGxKaI2AFcD5xcLBARP4yIrfnsvcDB/a2mmZlNhXOs0uccqzSUybGa\nBzxZmN8MHDdJ+dOA26dSKTMz6y/nWKXPOVZpKBNYRdkXk/R24CPAiT3XyMzMzCxRZQKrnwLzC/Pz\nyc5avUqesH41sCQinmv1QsVf+9Xr9S6qaWYpqNfr3rfNbKwpYvITUpJmAo8C7wCeAu4DlkfE+kKZ\nNwB3A6dGxJo2rxON95JEROz8226ZmaUt35816npMVfH4lapGftWKFStYvXo1S5d+jK1bV4+4Vv3U\n2MzS7qfJNPKrWl8SvJnFi7/OPffcPNxKTWO9Hr86nrGKiAlJZwF3AjOAayJivaTT8+evAj4B/Arw\nRUkAOyJiYbeVMTOzwXCOVfqcY5WGUuNYRcR3IuI3I+JNEXFpvuyqPKgiIj4aEb8aEcfkU1+CKg8U\namZmZimp9C1t/NNgMzMzS0mlAyszM+sPj2OVPo9jlYbK3SvQzMz6zzlW6XOOVRocWJmZJeqwwxbw\nxBMbelp31qzj+1wbMwMHVmZmydq+HWAd0Hz71rLrmlm/OcfKzGwMOD8nfe7DNPiMlZnZGHB+Tvrc\nh2lI4oyVx7MyMzOzFCQRWPknwmZmZpaCJAIrMzObGufnpM99mIbkcqxqtZovDZqZdcn5OelzH6Yh\nuTNWvixoZmZmVZVcYGVmZmZWVckGVr4caGZWnvNz0tepD7///VXsscfeXU+zZu3NbbfdNsSWTG+K\niOG8kRSN95JEROz8O9VlZlZN+X6qUddjqorHryqZN28BTz11E72MvD79NDaz6vXTcEwAvQ2nP3fu\nH7Ny5V+xbNmy/lYpcb0ev5I9Y1Xks1dmZjbeZgJ79zgl9zu2SusYWElaImmDpMcknd+mzOfz5x+W\ndEz/qzm5RkJ7FQKsYh0ajzstM+vGZNuOtyszsxGLiLYTMAPYCBwC7A6sBRY0lVkK3J4/Pg5Y0+a1\noqHxeJDLhm3FihWvqUPZujbWtfFV3AYaj1sti9i17bR6vmrbVV6fSY8zKUyjPLZM5qCDfitgXUB0\nnGq1WtRqtVJl05zIp1HXY3DToPpw7tz3xC233DLqzblyej1+Tf4kHA/cUZi/ALigqcyXgFMK8xuA\nA1q8VnNlY5DLhvGh0urDbir1b37Nyd6v1YevpafbgLzT8+3W6bVe7bavstufA6vB6iawmv7T9A+s\nBjU5sGptUIHV+4CrC/OnAl9oKrMKOKEw/z3gt1u8VnNlYxjLOml3VqDMOoOoa6uzEZOVaxWUlT37\n4aBsNKYSkPeyTi/bQ6cvK2W3PwdWg+XAqjg5sOp1cmDV2qACqz8qGVidWJj/HnBsi9dqrmwMY1mn\nb9xl1+n32aluPjQHFchV7cOiU6DX70BwsgCzl4C77PsNYhsZ9peVLtfp+sBUtalq+0qDA6vi5MCq\n18mBVWu9Hr8mHW5B0iKgFhFL8vkLgVci4lOFMl8C6hFxfT6/AXhbRGxpeq2AFYUli/PJzKaPej41\nXEx4uIWB6Wa4hcb4R9P3tijTf7iFQfXh3LnLWLnyox5uoUnPw8VMFnWR/QbzJ2TJ63vQOXl9ESWS\n1y09Zc/wdXsmcNj16pdW71dc5susGXr8xle1qarHL5+xKk4+Y9Xr5DNWrfV6/OpcAH4feJTs14EX\n5stOB04vlLkyf/5hWlwGzMsM4/9gA1Y2WBlkUFO19xtk4Jg6B1aD5cCqODmw6nVyYNVar8evkYy8\nbjad1Go1jx/VhkdeHyyPvF40/S8FDso++yzj+ONf4Y1vPKzrdffdd28+/elLB1Cr0ev1+OXAyswG\nxoFVZ5s3b2bTpk09rfve936Qn//8DpxjBeMQWA2uD1cBT/Sw3ovMmXM5L7zwTJ/rUw0OrMyschxY\ndXbZZZdx0UWXM3v2oV2vu2MHbNv2TbI02HE3/QOr6nmGOXOOcGDVxDcIMjMbsV/+cjlbt1426mqY\nWR9Mi5swm5mZmVWBAyszG3tlbjafulrt4p05OpYm92EaHFj1oF6vj7oKfTFd2gFui/VO0gyyIWOW\nAEcAyyVNu5/Z1Wor8qTn+qir0kf1UVegj+odS+zqw+ry8cs5Vj2p1+ssXrx41NWYsunSDnBbbEoW\nAhsjYhOApOuBk4H1ZV9g+/btXHfddT29+f333w/M72nd3tSZPne9qOO2jNbLL7/IOeect3N+zZof\nsGjRiaXWPemkt/Pudy8dVNVGxoGVmY27ecCThfnNwHHdvMBLL73Eaaf9JXvu+Sc9vP2eTEy8uYf1\nzEZtbyYm/o7Pfe7Vy+6999dKrHs3M2fOcGBlZjYN9en3+cGMGb/oac25c68Fru1PNdo499wseLvk\nkm8xe/Z/DvS9huXllx/d2Zbnn8+WzZ37nhHWqHfFtrTT6MPPfvaBYVSpJy+//D/Mnv3vHctt3/4E\n0lFDqNHwDXUcq6G8kZlVStXHsSp5s3kfv8zGUKUHCDUzqyJJM8nuh/oO4CngPmB5RJTOsTIza/Cl\nQDMbaxExIeks4E5gBnCNgyoz65XPWJmZmZn1yVDGsUp18D1J8yXdI+nHkv5L0l/ny18n6buS/lvS\nXZL2G3Vdy5I0Q9JDklbl80m2RdJ+km6UtF7SOknHpdgWSRfm29cjkr4laVYq7ZD0VUlbJD1SWNa2\n7nlbH8uPBe8aTa3ba7e/tyj3+bwdD0s6Ztj1LKNMWyQtlrQ1Px48JOmiUdS1E0mzJd0raW2+r1/a\nplyl+6VMO1Lpk4bmz5MWz1e6T4oma0vX/RIRA53ITq1vJLtL6O7AWmDBoN+3T3U/EDg6fzyHLA9j\nAfAPwN/my88HPjnqunbRpnPJfn50az6fZFuAbwAfyR/PBPZNrS35PvE4MCuf/2fgz1JpB/AW4Bjg\nkcKylnUnG3hzbX4MOCQ/Juw26jY0tafl/t5UZilwe/74OGDNqOs9hbYsbhwHqj4Be+V/ZwJrgN9N\ntF86tSOZPsnr+6rPkxT7pGRbuuqXYZyx2jn4XkTsABqD71VeRDwdEWvzxy+SDRg4D1hG9sFO/vcP\nR1PD7kg6mGxj/wq7bgWfXFsk7Qu8JSK+ClmOTERsJb22PA/sAPbKE6j3IkueTqIdEfEfwHNNi9vV\n/WTguojYEdlAnBvJjg2V0WZ/P6ip2M72RcS9wH6SDhhqRUso2RbYdRyotIjYlj/cg+zL+rNNRVLp\nl07tgET6pM3nSVESfQKl2sIky19jGIFVq8H35g3hfftK0iFk387vBQ6IiC35U1uASm4sLVwOnAe8\nUliWYlsOBZ6R9DVJD0q6WtLeJNaWiHgW+Azwv2QB1f9FxHdJrB1N2tX9ILJ9v6HSx4Gm/b2o1fHs\n4OHUqjeTtCWAE/LLNLdLOmLYdStL0m6S1pJtU/dExLqmIkn0S4l2JNMntP48KUqiT3Kd2tJVvwwj\nsEo+O17SHOBfgL+JiBeKz0V2nrDybZT0B8DPIuIh2kTeqbSF7DT6scA/RsSxwEvABcUCKbRF0q8D\n55BdGjsImCPp1GKZFNrRTom6V7Jd+f5+I9n+/mKrIk3zlWwHdGzLg8D8iDgK+AJw87DrV1ZEvBIR\nR5N9ML9V0uIWxSrfLyXakUSflPk8aRRtmq9cn5RsS1f9MozA6qe8+kZY83n1N9dKk7Q7WVC1MiIa\n/8wtkg7Mn3898LNR1a8LJwDLJD0BXAf8nqSVpNmWzcDmiLg/n7+RLNB6OrG2vBlYHRG/iIgJ4Cbg\neNJrR1G77an5OHBwvqxSCvv7Nwv7e1ES7YDObYmIFxqXpiLiO8Dukl435Gp2Jb/kfxvZvlOUTL9A\n+3Yk1CetPk/+qalMKn3SsS3d9sswAqsHgMMlHSJpD+AU4NYhvO+USRJwDbAuIq4oPHUrWZIx+d9K\nfqsoioiPR8T8iDgU+ABwd0T8KWm25WngSUm/kS96J/BjYBVptWUDsEjSnvm29k5gHem1o6jd9nQr\n8AFJe0g6FDicbCDOyphkfy+6FfhQXn4R2eXbLW3KjkyZtkg6IC+HpIVkw++0yvkZKUn7K/91qaQ9\ngZOAh5qKVb5fyrQjlT5p83nyoaZile8TKNeWbvtl4AOERtqD750InAr8SFJjB7gQ+CRwg6TTgE3A\n+0dTvSlpnJJNtS1nA9fmwfpPgA+TbV/JtCUiHs6/GT1Adm3/QeDLwD4k0A5J1wFvA/aX9CTwCdps\nTxGxTtINZIHjBHBmfqmwSlrt7x8H3gAQEVdFxO2SlkraSHYJ+sOjqWpHHdsCvA84Q9IEsI3sQ6WK\nXg98Q9JuZCcDVkbEv0k6HZLql47tIJ0+aRYACfZJK69pC132iwcINTMzM+uToQwQamZmZjYOHFiZ\nmZmZ9YkDKzMzM7M+cWBlZmZm1icOrMzMzMz6xIGVmZmZWZ84sDIzMzPrEwdWZmZmZn3y/yUP3s8f\nOjmCAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107e99f50>"
       ]
      }
     ],
     "prompt_number": 32
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The upper left-hand pane of each figure shows the temporal series of the\n",
      "samples from each parameter, while below is an autocorrelation plot of\n",
      "the samples. The right-hand pane shows a histogram of the trace. The\n",
      "trace is useful for evaluating and diagnosing the algorithm's\n",
      "performance, while the histogram is useful for\n",
      "visualizing the posterior.\n",
      "\n",
      "An alternative visualization of posterior quantities is to use the `summary_plot` function to create forest plots of stochastic or determinsitic nodes. This will display the posterior median, interquartile range, and posterior credible interval (defaults to 95%) in a layout that makes it easy to compare multiple nodes."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "Matplot.summary_plot([M.early_mean, M.late_mean])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Could not calculate Gelman-Rubin statistics. Requires multiple chains of equal length.\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAaQAAAEgCAYAAAD7bQf7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkCRJTTCQJElNMJAkSU34f4dhoMzDH1tPAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x1083289d0>"
       ]
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "For a non-graphical summary of the posterior, simply call the `summary` method."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M.early_mean.stats()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 34,
       "text": [
        "{'95% HPD interval': array([ 2.49124983,  3.60405374]),\n",
        " 'mc error': 0.00756500947969721,\n",
        " 'mean': 3.0527065227383434,\n",
        " 'n': 9000,\n",
        " 'quantiles': {2.5: 2.5172050574749441,\n",
        "  25: 2.8605823290604975,\n",
        "  50: 3.0380496362489353,\n",
        "  75: 3.2300775101774053,\n",
        "  97.5: 3.6363693972833957},\n",
        " 'standard deviation': 0.28702076399829468}"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### Imputation of Missing Data\n",
      "\n",
      "As with most textbook examples, the models we have examined so far\n",
      "assume that the associated data are complete. That is, there are no\n",
      "missing values corresponding to any observations in the dataset.\n",
      "However, many real-world datasets have missing observations, usually due\n",
      "to some logistical problem during the data collection process. The\n",
      "easiest way of dealing with observations that contain missing values is\n",
      "simply to exclude them from the analysis; this is called a **complete case analysis**. However, this results in loss\n",
      "of information if an excluded observation contains valid values for\n",
      "other quantities, and can bias results. An alternative is to ***impute*** the\n",
      "missing values, based on information in the rest of the model.\n",
      "\n",
      "For example, consider a survey dataset for some wildlife species:\n",
      "\n",
      "    Count   Site   Observer   Temperature\n",
      "    ------- ------ ---------- -------------\n",
      "    15      1      1          15\n",
      "    10      1      2          NA\n",
      "    6       1      1          11\n",
      "\n",
      "Each row contains the number of individuals seen during the survey,\n",
      "along with three covariates: the site on which the survey was conducted,\n",
      "the observer that collected the data, and the temperature during the\n",
      "survey. If we are interested in modelling, say, population size as a\n",
      "function of the count and the associated covariates, it is difficult to\n",
      "accommodate the second observation because the temperature is missing\n",
      "(perhaps the thermometer was broken that day). Ignoring this observation\n",
      "will allow us to fit the model, but it wastes information that is\n",
      "contained in the other covariates.\n",
      "\n",
      "In a Bayesian modelling framework, missing data are accommodated simply\n",
      "by treating them as unknown model parameters. Values for the missing\n",
      "data $\\tilde{y}$ are estimated naturally, using the posterior predictive\n",
      "distribution:\n",
      "\n",
      "<div style=\"font-size: 150%;\">  \n",
      "$$p(\\tilde{y}|y) = \\int p(\\tilde{y}|\\theta) f(\\theta|y) d\\theta$$\n",
      "</div>\n",
      "\n",
      "This describes additional data $\\tilde{y}$, which may either be\n",
      "considered unobserved data or potential future observations. We can use\n",
      "the posterior predictive distribution to model the likely values of\n",
      "missing data, assuming values are missing completely at random.\n",
      "\n",
      "Consider the coal mining disasters data introduced previously. Assume\n",
      "that two years of data are missing from the time series; we indicate\n",
      "this in the data array by the use of an arbitrary placeholder value,\n",
      "`None`:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "x = np.array([ 4, 5, 4, 0, 1, 4, 3, 4, 0, 6, 3, 3, 4, 0, 2, 6,\n",
      "3, 3, 5, 4, 5, 3, 1, 4, 4, 1, 5, 5, 3, 4, 2, 5,\n",
      "2, 2, 3, 4, 2, 1, 3, None, 2, 1, 1, 1, 1, 3, 0, 0,\n",
      "1, 0, 1, 1, 0, 0, 3, 1, 0, 3, 2, 2, 0, 1, 1, 1,\n",
      "0, 1, 0, 1, 0, 0, 0, 2, 1, 0, 0, 0, 1, 1, 0, 2,\n",
      "3, 3, 1, None, 2, 1, 1, 1, 1, 2, 4, 2, 0, 0, 1, 4,\n",
      "0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 35
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "To estimate these values in PyMC, we generate a *masked array*. These are specialised NumPy arrays that contain a matching True or False value for each element to indicate if that value should be excluded from any computation. Masked arrays can be generated using NumPy's `ma.masked_equal` function:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "masked_values = np.ma.masked_values(x, value=None)\n",
      "masked_values"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 36,
       "text": [
        "masked_array(data = [4 5 4 0 1 4 3 4 0 6 3 3 4 0 2 6 3 3 5 4 5 3 1 4 4 1 5 5 3 4 2 5 2 2 3 4 2\n",
        " 1 3 -- 2 1 1 1 1 3 0 0 1 0 1 1 0 0 3 1 0 3 2 2 0 1 1 1 0 1 0 1 0 0 0 2 1 0\n",
        " 0 0 1 1 0 2 3 3 1 -- 2 1 1 1 1 2 4 2 0 0 1 4 0 0 0 1 0 0 0 0 0 1 0 0 1 0 1],\n",
        "             mask = [False False False False False False False False False False False False\n",
        " False False False False False False False False False False False False\n",
        " False False False False False False False False False False False False\n",
        " False False False  True False False False False False False False False\n",
        " False False False False False False False False False False False False\n",
        " False False False False False False False False False False False False\n",
        " False False False False False False False False False False False  True\n",
        " False False False False False False False False False False False False\n",
        " False False False False False False False False False False False False\n",
        " False False False],\n",
        "       fill_value = ?)"
       ]
      }
     ],
     "prompt_number": 36
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This masked array, in turn, can then be passed to one of PyMC's data\n",
      "stochastic variables, which recognizes the masked array and replaces the\n",
      "missing values with Stochastic variables of the desired type. For the\n",
      "coal mining disasters problem, recall that disaster events were modeled\n",
      "as Poisson variates:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "disasters = Poisson('disasters', mu=rate, \n",
      "                    value=masked_values, observed=True)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 37
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here `rate` is an array of means for each year of data, allocated\n",
      "according to the location of the switchpoint. Each element in\n",
      "`disasters` is a Poisson Stochastic, irrespective of whether the\n",
      "observation was missing or not. The difference is that actual\n",
      "observations are data Stochastics (`observed=True`), while the missing\n",
      "values are non-data Stochastics. The latter are considered unknown,\n",
      "rather than fixed, and therefore estimated by the MCMC algorithm, just\n",
      "as unknown model parameters.\n",
      "\n",
      "The entire model looks very similar to the original model:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def missing_data_model():\n",
      "\n",
      "    # Switchpoint\n",
      "    switch = DiscreteUniform('switch', lower=0, upper=110)\n",
      "    # Early mean\n",
      "    early_mean = Exponential('early_mean', beta=1)\n",
      "    # Late mean\n",
      "    late_mean = Exponential('late_mean', beta=1)\n",
      "    \n",
      "    @deterministic(plot=False)\n",
      "    def rate(s=switch, e=early_mean, l=late_mean):\n",
      "        \"\"\"Allocate appropriate mean to time series\"\"\"\n",
      "        out = np.empty(len(disasters_array))\n",
      "        # Early mean prior to switchpoint\n",
      "        out[:s] = e\n",
      "        # Late mean following switchpoint\n",
      "        out[s:] = l\n",
      "        return out\n",
      "    \n",
      "    masked_values = np.ma.masked_values(x, value=None)\n",
      "    \n",
      "    # Pass masked array to data stochastic, and it does the right thing\n",
      "    disasters = Poisson('disasters', mu=rate, value=masked_values, observed=True)\n",
      "    \n",
      "    return locals()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 38
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here, we have used the `masked_values` function, rather than `masked_equal`; The result is the same."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M.early_mean.summary()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "early_mean:\n",
        " \n",
        "\tMean             SD               MC Error        95% HPD interval\n",
        "\t------------------------------------------------------------------\n",
        "\t3.053            0.287            0.008            [ 2.491  3.604]\n",
        "\t\n",
        "\t\n",
        "\tPosterior quantiles:\n",
        "\t\n",
        "\t2.5             25              50              75             97.5\n",
        "\t |---------------|===============|===============|---------------|\n",
        "\t2.517            2.861           3.038          3.23          3.636\n",
        "\t\n"
       ]
      }
     ],
     "prompt_number": 39
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M_missing = MCMC(missing_data_model())\n",
      "M_missing.sample(5000)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [--------------   38%                  ] 1946 of 5000 complete in 0.5 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------77%---------         ] 3862 of 5000 complete in 1.0 sec"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r",
        " [-----------------100%-----------------] 5000 of 5000 complete in 1.3 sec"
       ]
      }
     ],
     "prompt_number": 40
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "Matplot.plot(M_missing.disasters)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Plotting disasters_0\n",
        "Plotting"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " disasters_1\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAlEAAAFwCAYAAACYZC7QAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x107d68510>"
       ]
      }
     ],
     "prompt_number": 41
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Fine-tuning the MCMC algorithm\n",
      "\n",
      "MCMC objects handle individual variables via *step methods*, which\n",
      "determine how parameters are updated at each step of the MCMC algorithm.\n",
      "By default, step methods are automatically assigned to variables by\n",
      "PyMC. To see which step methods $M$ is using, look at its\n",
      "`step_method_dict` attribute with respect to each parameter:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "M.step_method_dict"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 42,
       "text": [
        "{<pymc.distributions.Exponential 'late_mean' at 0x107a034d0>: [<pymc.StepMethods.Metropolis at 0x107a2f710>],\n",
        " <pymc.distributions.Exponential 'early_mean' at 0x107a035d0>: [<pymc.StepMethods.Metropolis at 0x107a2f750>],\n",
        " <pymc.distributions.DiscreteUniform 'switchpoint' at 0x107a13810>: [<pymc.StepMethods.DiscreteMetropolis at 0x107a2fc10>]}"
       ]
      }
     ],
     "prompt_number": 42
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The value of `step_method_dict` corresponding to a particular variable\n",
      "is a list of the step methods $M$ is using to handle that variable.\n",
      "\n",
      "You can force $M$ to use a particular step method by calling `M.use_step_method` before telling it to sample. The following call will cause $M$ to handle `late_mean` with a `Slicer` (slice sampling) step method, and assigns an initial slice width `w=0.5`:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from pymc import Slicer\n",
      "M.use_step_method(Slicer, disaster_model.late_mean, w=0.5)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 43
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Another step method class, `AdaptiveMetropolis`, is better at handling\n",
      "highly-correlated variables. If your model mixes poorly, using\n",
      "`AdaptiveMetropolis` is a sensible first thing to try."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "---"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.core.display import HTML\n",
      "def css_styling():\n",
      "    styles = open(\"styles/custom.css\", \"r\").read()\n",
      "    return HTML(styles)\n",
      "css_styling()\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<style>\n",
        "    @font-face {\n",
        "        font-family: \"Computer Modern\";\n",
        "        src: url('http://mirrors.ctan.org/fonts/cm-unicode/fonts/otf/cmunss.otf');\n",
        "    }\n",
        "    div.cell{\n",
        "        width: 90%;\n",
        "/*        margin-left:auto;*/\n",
        "/*        margin-right:auto;*/\n",
        "    }\n",
        "    ul {\n",
        "        line-height: 145%;\n",
        "        font-size: 90%;\n",
        "    }\n",
        "    li {\n",
        "        margin-bottom: 1em;\n",
        "    }\n",
        "    h1 {\n",
        "        font-family: Helvetica, serif;\n",
        "    }\n",
        "    h4{\n",
        "        margin-top: 12px;\n",
        "        margin-bottom: 3px;\n",
        "       }\n",
        "    div.text_cell_render{\n",
        "        font-family: Computer Modern, \"Helvetica Neue\", Arial, Helvetica, Geneva, sans-serif;\n",
        "        line-height: 145%;\n",
        "        font-size: 130%;\n",
        "        width: 90%;\n",
        "        margin-left:auto;\n",
        "        margin-right:auto;\n",
        "    }\n",
        "    .CodeMirror{\n",
        "            font-family: \"Source Code Pro\", source-code-pro,Consolas, monospace;\n",
        "    }\n",
        "/*    .prompt{\n",
        "        display: None;\n",
        "    }*/\n",
        "    .text_cell_render h5 {\n",
        "        font-weight: 300;\n",
        "        font-size: 16pt;\n",
        "        color: #4057A1;\n",
        "        font-style: italic;\n",
        "        margin-bottom: 0.5em;\n",
        "        margin-top: 0.5em;\n",
        "        display: block;\n",
        "    }\n",
        "\n",
        "    .warning{\n",
        "        color: rgb( 240, 20, 20 )\n",
        "        }\n",
        "</style>\n",
        "<script>\n",
        "    MathJax.Hub.Config({\n",
        "                        TeX: {\n",
        "                           extensions: [\"AMSmath.js\"]\n",
        "                           },\n",
        "                tex2jax: {\n",
        "                    inlineMath: [ ['$','$'], [\"\\\\(\",\"\\\\)\"] ],\n",
        "                    displayMath: [ ['$$','$$'], [\"\\\\[\",\"\\\\]\"] ]\n",
        "                },\n",
        "                displayAlign: 'center', // Change this to 'center' to center equations.\n",
        "                \"HTML-CSS\": {\n",
        "                    styles: {'.MathJax_Display': {\"margin\": 4}}\n",
        "                }\n",
        "        });\n",
        "</script>\n"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 44,
       "text": [
        "<IPython.core.display.HTML object>"
       ]
      }
     ],
     "prompt_number": 44
    }
   ],
   "metadata": {}
  }
 ]
}