{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chapter 15. Missing Data and Other Opportunities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install -q numpyro arviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "import os\n",
    "\n",
    "import arviz as az\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import jax.numpy as jnp\n",
    "from jax import ops, random, vmap\n",
    "from jax.scipy.special import expit\n",
    "\n",
    "import numpyro\n",
    "import numpyro.distributions as dist\n",
    "from numpyro.diagnostics import print_summary\n",
    "from numpyro.distributions import constraints\n",
    "from numpyro.infer import MCMC, NUTS, init_to_value\n",
    "\n",
    "if \"SVG\" in os.environ:\n",
    "    %config InlineBackend.figure_formats = [\"svg\"]\n",
    "az.style.use(\"arviz-darkgrid\")\n",
    "numpyro.set_platform(\"cpu\")\n",
    "numpyro.set_host_device_count(4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DeviceArray(0.6641716, dtype=float32)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# simulate a pancake and return randomly ordered sides\n",
    "def sim_pancake(seed):\n",
    "    pancake = dist.Categorical(logits=jnp.ones(3)).sample(random.PRNGKey(2 * seed))\n",
    "    sides = jnp.array([1, 1, 1, 0, 0, 0]).reshape(3, 2).T[:, pancake]\n",
    "    return random.permutation(random.PRNGKey(2 * seed + 1), sides)\n",
    "\n",
    "\n",
    "# sim 10,000 pancakes\n",
    "pancakes = vmap(sim_pancake, out_axes=1)(jnp.arange(10000))\n",
    "up = pancakes[0]\n",
    "down = pancakes[1]\n",
    "\n",
    "# compute proportion 1/1 (BB) out of all 1/1 and 1/0\n",
    "num_11_10 = jnp.sum(up == 1)\n",
    "num_11 = jnp.sum((up == 1) & (down == 1))\n",
    "num_11 / num_11_10"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "WaffleDivorce = pd.read_csv(\"../data/WaffleDivorce.csv\", sep=\";\")\n",
    "d = WaffleDivorce\n",
    "\n",
    "# points\n",
    "ax = az.plot_pair(\n",
    "    d[[\"MedianAgeMarriage\", \"Divorce\"]].to_dict(orient=\"list\"),\n",
    "    scatter_kwargs=dict(s=15, facecolors=\"none\"),\n",
    ")\n",
    "ax.set(ylim=(4, 15), xlabel=\"Median age marrage\", ylabel=\"Divorce rate\")\n",
    "\n",
    "# standard errors\n",
    "for i in range(d.shape[0]):\n",
    "    ci = d.Divorce[i] + jnp.array([-1, 1]) * d[\"Divorce SE\"][i]\n",
    "    x = d.MedianAgeMarriage[i]\n",
    "    plt.plot([x, x], ci, \"k\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "586c6ba23f194b6f9c8f383421399ec8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2a869125e6354c70a2bd6656952a55f3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "afd225f9523147ad84577b03d2ba97f8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e172ff3ec244a29baf69b4132c154df",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dlist = dict(\n",
    "    D_obs=d.Divorce.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    D_sd=d[\"Divorce SE\"].values / d.Divorce.std(),\n",
    "    M=d.Marriage.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    A=d.MedianAgeMarriage.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    N=d.shape[0],\n",
    ")\n",
    "\n",
    "\n",
    "def model(A, M, D_sd, D_obs, N):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 0.2))\n",
    "    bA = numpyro.sample(\"bA\", dist.Normal(0, 0.5))\n",
    "    bM = numpyro.sample(\"bM\", dist.Normal(0, 0.5))\n",
    "    sigma = numpyro.sample(\"sigma\", dist.Exponential(1))\n",
    "    mu = a + bA * A + bM * M\n",
    "    D_true = numpyro.sample(\"D_true\", dist.Normal(mu, sigma))\n",
    "    numpyro.sample(\"D_obs\", dist.Normal(D_true, D_sd), obs=D_obs)\n",
    "\n",
    "\n",
    "m15_1 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_1.run(random.PRNGKey(0), **dlist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                mean       std    median      5.5%     94.5%     n_eff     r_hat\n",
      " D_true[0]      1.16      0.36      1.17      0.61      1.75   3190.26      1.00\n",
      " D_true[1]      0.70      0.52      0.69     -0.14      1.50   3050.85      1.00\n",
      " D_true[2]      0.42      0.32      0.42     -0.05      0.97   4423.54      1.00\n",
      " D_true[3]      1.42      0.47      1.41      0.62      2.13   4693.23      1.00\n",
      " D_true[4]     -0.90      0.13     -0.90     -1.09     -0.68   4757.64      1.00\n",
      " D_true[5]      0.65      0.38      0.64      0.04      1.27   3480.80      1.00\n",
      " D_true[6]     -1.36      0.35     -1.36     -1.89     -0.77   3693.32      1.00\n",
      " D_true[7]     -0.32      0.48     -0.32     -1.08      0.45   3808.76      1.00\n",
      " D_true[8]     -1.88      0.60     -1.88     -2.85     -0.93   3263.41      1.00\n",
      " D_true[9]     -0.62      0.17     -0.62     -0.87     -0.34   5084.32      1.00\n",
      "D_true[10]      0.76      0.28      0.76      0.32      1.19   4423.47      1.00\n",
      "D_true[11]     -0.54      0.49     -0.54     -1.33      0.18   3491.72      1.00\n",
      "D_true[12]      0.19      0.49      0.20     -0.59      0.99   2026.07      1.00\n",
      "D_true[13]     -0.87      0.22     -0.88     -1.23     -0.55   5782.09      1.00\n",
      "D_true[14]      0.56      0.31      0.55      0.11      1.12   4043.40      1.00\n",
      "D_true[15]      0.28      0.37      0.28     -0.34      0.83   4969.48      1.00\n",
      "D_true[16]      0.49      0.39      0.49     -0.11      1.14   3902.02      1.00\n",
      "D_true[17]      1.25      0.34      1.25      0.68      1.77   4268.92      1.00\n",
      "D_true[18]      0.43      0.39      0.42     -0.20      1.04   3958.69      1.00\n",
      "D_true[19]      0.39      0.53      0.38     -0.42      1.25   2242.96      1.00\n",
      "D_true[20]     -0.55      0.33     -0.56     -1.06     -0.03   4700.87      1.00\n",
      "D_true[21]     -1.11      0.25     -1.10     -1.49     -0.70   3459.65      1.00\n",
      "D_true[22]     -0.27      0.25     -0.27     -0.66      0.12   3367.35      1.00\n",
      "D_true[23]     -1.00      0.30     -1.00     -1.47     -0.54   3468.74      1.00\n",
      "D_true[24]      0.43      0.41      0.43     -0.23      1.06   4192.29      1.00\n",
      "D_true[25]     -0.03      0.31     -0.02     -0.51      0.49   4341.29      1.00\n",
      "D_true[26]     -0.03      0.49     -0.02     -0.81      0.71   4395.28      1.00\n",
      "D_true[27]     -0.14      0.38     -0.15     -0.77      0.43   4835.72      1.00\n",
      "D_true[28]     -0.25      0.49     -0.27     -0.99      0.56   3569.33      1.00\n",
      "D_true[29]     -1.79      0.24     -1.80     -2.16     -1.42   4291.70      1.00\n",
      "D_true[30]      0.18      0.40      0.17     -0.43      0.83   4206.43      1.00\n",
      "D_true[31]     -1.66      0.16     -1.66     -1.92     -1.40   5105.78      1.00\n",
      "D_true[32]      0.12      0.23      0.12     -0.24      0.48   5028.23      1.00\n",
      "D_true[33]     -0.05      0.53     -0.04     -0.87      0.77   3471.07      1.00\n",
      "D_true[34]     -0.12      0.23     -0.12     -0.48      0.23   4786.81      1.00\n",
      "D_true[35]      1.28      0.40      1.28      0.61      1.89   3415.97      1.00\n",
      "D_true[36]      0.23      0.34      0.24     -0.34      0.75   4679.23      1.00\n",
      "D_true[37]     -1.03      0.22     -1.02     -1.36     -0.65   4103.00      1.00\n",
      "D_true[38]     -0.91      0.54     -0.92     -1.84     -0.13   3537.37      1.00\n",
      "D_true[39]     -0.67      0.31     -0.67     -1.16     -0.18   3980.07      1.00\n",
      "D_true[40]      0.24      0.56      0.24     -0.60      1.16   4535.89      1.00\n",
      "D_true[41]      0.74      0.33      0.74      0.20      1.23   4946.62      1.00\n",
      "D_true[42]      0.20      0.18      0.20     -0.09      0.47   4276.96      1.00\n",
      "D_true[43]      0.81      0.44      0.82      0.12      1.49   2592.55      1.00\n",
      "D_true[44]     -0.42      0.53     -0.43     -1.23      0.44   3480.32      1.00\n",
      "D_true[45]     -0.39      0.25     -0.39     -0.78      0.01   5468.52      1.00\n",
      "D_true[46]      0.14      0.29      0.14     -0.36      0.58   4116.98      1.00\n",
      "D_true[47]      0.56      0.46      0.56     -0.13      1.33   3906.03      1.00\n",
      "D_true[48]     -0.63      0.28     -0.64     -1.05     -0.16   4596.77      1.00\n",
      "D_true[49]      0.86      0.59      0.86     -0.05      1.84   3001.88      1.00\n",
      "         a     -0.05      0.09     -0.06     -0.19      0.10   3053.48      1.00\n",
      "        bA     -0.61      0.16     -0.62     -0.84     -0.34   2406.74      1.00\n",
      "        bM      0.05      0.17      0.05     -0.23      0.30   2181.48      1.00\n",
      "     sigma      0.58      0.10      0.57      0.41      0.74    832.29      1.00\n",
      "\n",
      "Number of divergences: 0\n"
     ]
    }
   ],
   "source": [
    "m15_1.print_summary(0.89)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d80e47e1a024991b98bb4e8e6e8f1e2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "25821756959e40dfa29ba30439cde2fa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d4f1a588621456486c211d234635408",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "28d4a96de0934319a566842ace182fc3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dlist = dict(\n",
    "    D_obs=d.Divorce.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    D_sd=d[\"Divorce SE\"].values / d.Divorce.std(),\n",
    "    M_obs=d.Marriage.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    M_sd=d[\"Marriage SE\"].values / d.Marriage.std(),\n",
    "    A=d.MedianAgeMarriage.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    N=d.shape[0],\n",
    ")\n",
    "\n",
    "\n",
    "def model(A, M_sd, M_obs, D_sd, D_obs, N):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 0.2))\n",
    "    bA = numpyro.sample(\"bA\", dist.Normal(0, 0.5))\n",
    "    bM = numpyro.sample(\"bM\", dist.Normal(0, 0.5))\n",
    "    sigma = numpyro.sample(\"sigma\", dist.Exponential(1))\n",
    "    M_est = numpyro.sample(\"M_est\", dist.Normal(0, 1).expand([N]))\n",
    "    numpyro.sample(\"M_obs\", dist.Normal(M_est, M_sd), obs=M_obs)\n",
    "    mu = a + bA * A + bM * M_est\n",
    "    D_est = numpyro.sample(\"D_est\", dist.Normal(mu, sigma))\n",
    "    numpyro.sample(\"D_obs\", dist.Normal(D_est, D_sd), obs=D_obs)\n",
    "\n",
    "\n",
    "m15_2 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_2.run(random.PRNGKey(0), **dlist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "post = m15_2.get_samples()\n",
    "D_est = jnp.mean(post[\"D_est\"], 0)\n",
    "M_est = jnp.mean(post[\"M_est\"], 0)\n",
    "plt.plot(dlist[\"M_obs\"], dlist[\"D_obs\"], \"bo\", alpha=0.5)\n",
    "plt.gca().set(xlabel=\"marriage rate (std)\", ylabel=\"divorce rate (std)\")\n",
    "plt.plot(M_est, D_est, \"ko\", mfc=\"none\")\n",
    "for i in range(d.shape[0]):\n",
    "    plt.plot([dlist[\"M_obs\"][i], M_est[i]], [dlist[\"D_obs\"][i], D_est[i]], \"k-\", lw=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.7"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "N = 500\n",
    "A = dist.Normal().sample(random.PRNGKey(0), (N,))\n",
    "M = dist.Normal(-A).sample(random.PRNGKey(1))\n",
    "D = dist.Normal(A).sample(random.PRNGKey(2))\n",
    "A_obs = dist.Normal(A).sample(random.PRNGKey(3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "N = 100\n",
    "S = dist.Normal().sample(random.PRNGKey(0), (N,))\n",
    "H = dist.Binomial(10, expit(S)).sample(random.PRNGKey(1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "D = dist.Bernoulli(0.5).sample(random.PRNGKey(2), (N,))  # dogs completely random\n",
    "Hm = jnp.where(D == 1, jnp.nan, H)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "D = jnp.where(S > 0, 1, 0)\n",
    "Hm = jnp.where(D == 1, jnp.nan, H)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.11"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "with numpyro.handlers.seed(rng_seed=501):\n",
    "    N = 1000\n",
    "    X = numpyro.sample(\"X\", dist.Normal().expand([N]))\n",
    "    S = numpyro.sample(\"S\", dist.Normal().expand([N]))\n",
    "    H = numpyro.sample(\"H\", dist.Binomial(10, logits=2 + S - 2 * X))\n",
    "    D = jnp.where(X > 1, 1, 0)\n",
    "    Hm = jnp.where(D == 1, jnp.nan, H)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.12"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cbe1c515354d4712a6573033fc722bac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6b6231a9386943cbb1c934f660346a55",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "06b3152c23894076be1bc124fad7800c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00000ccc705a4c0d95afd6e97f565af4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                mean       std    median      5.0%     95.0%     n_eff     r_hat\n",
      "         a      1.32      0.03      1.32      1.28      1.36   1240.12      1.00\n",
      "        bS      0.62      0.03      0.62      0.58      0.67   1298.89      1.00\n",
      "\n",
      "Number of divergences: 0\n"
     ]
    }
   ],
   "source": [
    "dat_list = dict(H=H, S=S)\n",
    "\n",
    "\n",
    "def model(S, H):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 1))\n",
    "    bS = numpyro.sample(\"bS\", dist.Normal(0, 0.5))\n",
    "    logit_p = a + bS * S\n",
    "    numpyro.sample(\"H\", dist.Binomial(10, logits=logit_p), obs=H)\n",
    "\n",
    "\n",
    "m15_3 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_3.run(random.PRNGKey(0), **dat_list)\n",
    "m15_3.print_summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.13"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "001e3c4172664103bd5ce945bd94bcca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2981d1907fb6496b9a6d44bdcfa45da1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1851b89791584849aef2f042d0611b3c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1990acabd4c24c36a752d68d5379e30b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                mean       std    median      5.0%     95.0%     n_eff     r_hat\n",
      "         a      1.92      0.03      1.92      1.86      1.97   1023.26      1.00\n",
      "        bS      0.73      0.03      0.73      0.67      0.78    989.63      1.00\n",
      "\n",
      "Number of divergences: 0\n"
     ]
    }
   ],
   "source": [
    "dat_list0 = dict(H=H[D == 0], S=S[D == 0])\n",
    "\n",
    "\n",
    "def model(S, H):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 1))\n",
    "    bS = numpyro.sample(\"bS\", dist.Normal(0, 0.5))\n",
    "    logit_p = a + bS * S\n",
    "    numpyro.sample(\"H\", dist.Binomial(10, logits=logit_p), obs=H)\n",
    "\n",
    "\n",
    "m15_4 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_4.run(random.PRNGKey(0), **dat_list0)\n",
    "m15_4.print_summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.14"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "D = jnp.where(jnp.abs(X) < 1, 1, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "N = 100\n",
    "S = dist.Normal().sample(random.PRNGKey(0), (N,))\n",
    "H = dist.Binomial(10, logits=S).sample(random.PRNGKey(1))\n",
    "D = jnp.where(H < 5, 1, 0)\n",
    "Hm = jnp.where(D == 1, jnp.nan, H)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "milk = pd.read_csv(\"../data/milk.csv\", sep=\";\")\n",
    "d = milk\n",
    "d[\"neocortex.prop\"] = d[\"neocortex.perc\"] / 100\n",
    "d[\"logmass\"] = d.mass.apply(math.log)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.17"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7eb1d7d3d6cb4c2e83889085161c89b5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d1fcaec632b41849e2a0e48c405a9ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "38c0195ea64e48aca29e7fb28d988fe4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2c04767b346e46f4b012f31b6b5fb6a2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dat_list = dict(\n",
    "    K=d[\"kcal.per.g\"].pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    B=d[\"neocortex.prop\"].pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    "    M=d.logmass.pipe(lambda x: (x - x.mean()) / x.std()).values,\n",
    ")\n",
    "\n",
    "\n",
    "def model(B, M, K):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 0.5))\n",
    "    nu = numpyro.sample(\"nu\", dist.Normal(0, 0.5))\n",
    "    bB = numpyro.sample(\"bB\", dist.Normal(0, 0.5))\n",
    "    bM = numpyro.sample(\"bM\", dist.Normal(0, 0.5))\n",
    "    sigma_B = numpyro.sample(\"sigma_B\", dist.Exponential(1))\n",
    "    sigma = numpyro.sample(\"sigma\", dist.Exponential(1))\n",
    "    B_impute = numpyro.sample(\n",
    "        \"B_impute\", dist.Normal(0, 1).expand([int(np.isnan(B).sum())]).mask(False)\n",
    "    )\n",
    "    B = jnp.asarray(B).at[np.nonzero(np.isnan(B))[0]].set(B_impute)\n",
    "    numpyro.sample(\"B\", dist.Normal(nu, sigma_B), obs=B)\n",
    "    mu = a + bB * B + bM * M\n",
    "    numpyro.sample(\"K\", dist.Normal(mu, sigma), obs=K)\n",
    "\n",
    "\n",
    "m15_5 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_5.run(random.PRNGKey(0), **dat_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.18"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                  mean       std    median      5.5%     94.5%     n_eff     r_hat\n",
      " B_impute[0]     -0.50      0.91     -0.54     -1.90      0.95   1918.23      1.00\n",
      " B_impute[1]     -0.64      0.92     -0.65     -2.15      0.74   1424.06      1.00\n",
      " B_impute[2]     -0.68      0.94     -0.71     -2.27      0.78   1338.37      1.00\n",
      " B_impute[3]     -0.26      0.88     -0.27     -1.65      1.09   2203.85      1.00\n",
      " B_impute[4]      0.43      0.88      0.42     -0.87      1.95   1664.89      1.00\n",
      " B_impute[5]     -0.15      0.90     -0.19     -1.57      1.29   1910.72      1.00\n",
      " B_impute[6]      0.17      0.92      0.18     -1.30      1.62   2257.71      1.00\n",
      " B_impute[7]      0.24      0.87      0.25     -0.99      1.80   2024.09      1.00\n",
      " B_impute[8]      0.49      0.92      0.49     -0.87      2.05   1943.39      1.00\n",
      " B_impute[9]     -0.42      0.93     -0.42     -1.93      0.95   1966.23      1.00\n",
      "B_impute[10]     -0.26      0.92     -0.27     -1.70      1.21   1980.72      1.00\n",
      "B_impute[11]      0.15      0.95      0.17     -1.36      1.64   1693.70      1.00\n",
      "           a      0.02      0.17      0.03     -0.24      0.29   1930.84      1.00\n",
      "          bB      0.49      0.24      0.50      0.09      0.86    775.56      1.00\n",
      "          bM     -0.54      0.21     -0.55     -0.86     -0.20    890.11      1.00\n",
      "          nu     -0.04      0.21     -0.04     -0.36      0.29   1704.73      1.00\n",
      "       sigma      0.85      0.14      0.84      0.64      1.07   1021.42      1.00\n",
      "     sigma_B      1.01      0.17      0.99      0.75      1.26   1246.13      1.00\n",
      "\n",
      "Number of divergences: 0\n"
     ]
    }
   ],
   "source": [
    "m15_5.print_summary(0.89)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51c833aaac334e61ac476e6c250afac3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dff93d80cf8a46e19c895cc2e9ecb9b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2ff8fee9c043415ebf94a61e769efddf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e192a7097d544dd8913237070cca574",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                mean       std    median      5.5%     94.5%     n_eff     r_hat\n",
      "         a      0.10      0.19      0.11     -0.18      0.42   1803.62      1.00\n",
      "        bB      0.60      0.28      0.61      0.17      1.04   1426.34      1.00\n",
      "        bM     -0.63      0.25     -0.65     -1.06     -0.28   1259.25      1.00\n",
      "        nu     -0.00      0.23     -0.00     -0.34      0.37   2292.07      1.00\n",
      "     sigma      0.87      0.18      0.85      0.59      1.12   1300.49      1.00\n",
      "   sigma_B      1.03      0.18      1.01      0.77      1.31   1956.68      1.00\n",
      "\n",
      "Number of divergences: 0\n"
     ]
    }
   ],
   "source": [
    "obs_idx = d[\"neocortex.prop\"].notnull().values\n",
    "dat_list_obs = dict(\n",
    "    K=dat_list[\"K\"][obs_idx], B=dat_list[\"B\"][obs_idx], M=dat_list[\"M\"][obs_idx]\n",
    ")\n",
    "\n",
    "\n",
    "def model(B, M, K):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 0.5))\n",
    "    nu = numpyro.sample(\"nu\", dist.Normal(0, 0.5))\n",
    "    bB = numpyro.sample(\"bB\", dist.Normal(0, 0.5))\n",
    "    bM = numpyro.sample(\"bM\", dist.Normal(0, 0.5))\n",
    "    sigma_B = numpyro.sample(\"sigma_B\", dist.Exponential(1))\n",
    "    sigma = numpyro.sample(\"sigma\", dist.Exponential(1))\n",
    "    numpyro.sample(\"B\", dist.Normal(nu, sigma_B), obs=B)\n",
    "    mu = a + bB * B + bM * M\n",
    "    numpyro.sample(\"K\", dist.Normal(mu, sigma), obs=K)\n",
    "\n",
    "\n",
    "m15_6 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_6.run(random.PRNGKey(0), **dat_list_obs)\n",
    "m15_6.print_summary(0.89)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x740 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "az.plot_forest(\n",
    "    [az.from_numpyro(m15_5), az.from_numpyro(m15_6)],\n",
    "    model_names=[\"m15.5\", \"m15.6\"],\n",
    "    var_names=[\"bB\", \"bM\"],\n",
    "    combined=True,\n",
    "    hdi_prob=0.89,\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.21"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtsAAAHrCAYAAAAe4lGYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAA9hAAAPYQGoP6dpAABg8ElEQVR4nO3de1yUdd7/8fcwYN54IESwRIXagkRFg8RM1DJTU3GtVat7Scm1w5aW9jPTLb03rdWy1NKOthuWtZtk1o1bYkfNPCDiMVFbCzFUJEXASIHh+v3BzSRycBxmmAOv5+Oxj43r+s51fQa/Xr7nO9/re5kMwzAEAAAAwOF8XF0AAAAA4K0I2wAAAICTELYBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAADgJL6uLsDbFRQUuLoEpwkICFBhYaGry4Cbo5/AVvQV2IJ+Als1Rl8JDAy8YBtGtmE3Hx+6Dy6MfgJb0VdgC/oJbOUufcU9qgAAAAC8EGEbAAAAcBLCNgAAAOAkhG0AAADASQjbAAAAgJMQtgEAAAAnIWwDAAAATkLYBgAAAJyEsA0AAAA4CWEbAAAAcBLCNgAAAOAkhG0AAADASQjbAAAAgJP4uroAAAAgWSyGdu6STpyUgtpI3aMls9nk6rIANBBhGwAAF1u33tCixYby83/bFhwsTZ4k9e9H4AY8GdNIAABwoXXrDT0xq3rQlqT8fOmJWYbWrTdcUxgAh/DKke28vDx9+umnWr9+vX744Qf9/PPPCggIUExMjCZMmKDu3bvbfKyKigq9++67WrFihQ4dOiR/f3/dcMMNmjJlijp27OjEdwEA8HYWS+WIdn1eXGIovg9TSgBP5ZUj2++8847mzp2rw4cPq0+fPrrnnnsUGxurL774Qnfeeac++eQTm481a9YsPf300zIMQ3fffbf69u2rtWvXatSoUcrOznbemwAAeL2du1RjRPt8x49XtgPgmbxyZDs6OlrvvPOO4uLiqm3PyMhQUlKS/vrXv2rgwIFq1qxZvcfZvHmzUlJS1LNnT/3jH/+wth8+fLjuu+8+zZkzR3//+9+d9j4AAN7txEnHtgPgfrxyZHvQoEE1grYkXXfdderVq5cKCwu1f//+Cx4nJSVFkvTII49UC+b9+/dXXFycNmzYoCNHjjiucABAkxLUxrHtALgfrwzb9fH19a32//XZsmWL/P39FRMTU2Nf3759JUnp6emOLRAA0GR0j65cdaQ+ISGV7QB4piYVto8cOaKNGzcqODhYERER9bYtKSlRfn6+OnToILPZXGN/WFiYJOnQoUNOqRUA4P3MZpMmT6r/xsdHJpq4ORLwYF45Z7s2ZWVlmjZtmkpLSzV16tRaA/S5iouLJUktW7asdX/V9qp2dQkICJCPj/d+pgkMDHR1CfAA9BPYqin2lZG/l1q0OKu5z5UoL6/Cuv2ydj6aPs1ftwy8xIXVuaem2E9gH3foK00ibFdUVGj69OnaunWrxowZo5EjRzbauQsLCxvtXI0tMDBQBQUFri4Dbo5+Als15b5yXay04j1DO3eZznmCpCGzuUQFBSWuLs+tNOV+govTGH3FljDv9WG7oqJCf/nLX7R69WqNGDFCTz31lE2va9WqlSTp9OnTte6v2l7VDgCAhjCbTYq51tVVAHA0753foMqgPWPGDK1atUrDhw/XvHnzbJ7S4e/vr+DgYP3000+yWCw19lfN1a6auw0AAACcz2vDdlXQ/uijjzR06FA999xzF5ynfb64uDiVlJQoMzOzxr5vvvlGktSzZ0+H1AsAAADv45Vhu2rqyEcffaQhQ4Zo/vz59QbtkydP6uDBgzp5svpTA8aMGSNJevHFF1VaWmrdvm7dOqWnpys+Pl6hoaHOeRMAAADweF45Z/vll1/WqlWr5O/vr/DwcL366qs12gwcOFCdO3eWJL377rtasmSJJk6cqEmTJlnbXH/99Ro9erRSUlJ0++23q3///srPz9cnn3yiSy+9VE8++WSjvScAAAB4Hq8M27m5uZIq18p+7bXXam0TGhpqDdv1mT17tiIiIrRixQq9/fbb8vf31y233KIpU6aoU6dODq0bAAAA3sVkGIbh6iK8mTcvT8TyS7AF/QS2oq/AFvQT2Mpdlv7zyjnbAAAAgDsgbAMAAABOQtgGAAAAnISwDQAAADgJYRsAAABwEsI2AAAA4CSEbQAAAMBJCNsAAACAkxC2AQAAACchbAMAAABOQtgGAAAAnISwDQAAADgJYRsAAABwEsI2AAAA4CSEbQAAAMBJCNsAAACAkxC2AQAAACchbAMAAABOQtgGAAAAnMTX1QUAANDYLBZDO3dJJ05KQW2k7tGS2WxydVkAvBBhGwDQpKxbb2jRYkP5+b9tCw6WJk+S+vcjcANwLKaRAACajHXrDT0xq3rQlqT8fOmJWYbWrTdcUxgAr0XYBgA0CRZL5Yh2fV5cYshiIXADcBzCNgCgSdi5SzVGtM93/HhlOwBwFMI2AKBJOHHSse0AwBaEbQBAkxDUxrHtAMAWhG0AQJPQPbpy1ZH6hIRUtgMARyFsAwCaBLPZpMmT6l/a75GJJqevt22xWLRhwwatXLlSGzZskMVicer5ALgWYRsA0GT072fSM7NNNUa4Q0KkZ2abnL7OdmpqqmJjYzVixAjde++9GjFihGJjY5WamurU8wJwHR5qAwBoUvr3Mym+jxr9CZKpqalKSkrS4MGDtXTpUnXu3FlZWVlauHChkpKSlJycrISEBKfWAKDxmQzDYEFRJyooKHB1CU4TGBjo1e8PjkE/ga28ua9YLBbFxsYqKipKy5cvl4/Pb18sV1RUKDExUVlZWcrIyJDZbHZhpe7Pm/sJHKsx+kpgYOAF2zCyDcClsrOzVVhY6Ooy4AZat26toqIiV5fhFDt27FBOTo6mTZum3bt319g/bNgwrVmzRsuXL1ePHj0av0AP4s39BDUFBAQoPDzc1WU0CCPbTubNn74ZXYAt6usnJ06cUGRkpCoqKhq5KgCAJzCbzdq3b5+CgoIu+rWMbANo8oKCgpSRkcHINiR594jljh07NGXKFC1ZskRdunSpsX/Pnj2aNGmSFi5cyMj2BXhzP0FNAQEBdgVtd8LItpN588gvI9uwBf0EtvLmvsKcbcfx5n4Cx3KXkW2W/gMAwMnMZrPmzJmjtLQ0JSYmKj09XcXFxUpPT1diYqLS0tI0e/ZsgjbghZhGAgBAI0hISFBycrJmzpypIUOGWLeHhYWx7B/gxQjbAAA0koSEBA0dOlSbNm1SXl6e2rVrp969ezOiDXgxwjYAAI3IbDYrPj7e1WUAaCTM2QYAAACchLANAAAAOAlhGwAAAHAS5mwDAFzOYjGUvrVM2YcMBbWRukdLZrPJ1WUBQIMRtgEALrVuvaFFiw3l5//2VMDgYGnyJKl/PwI3AM/GNBIAgMusW2/oiVmG8vOrb8/Pl56YZWjdeh5yDMCzEbYBAC5hsVSOaNfnxSWGLBYCNwDP5bXTSD7++GNt27ZNe/bs0YEDB1RWVqa5c+fq9ttvt/kYW7Zs0dixY+vcf7HHAwD8Zucu1RjRPt/x45XtYq5tnJoAwNG8Nmy/+OKLys3NVWBgoEJCQpSbm2v3seLi4hQXF1dje+fOnRtSIgA0aSdOOrYdALgjrw3bTz/9tMLCwhQaGqo33nhDL7zwgt3HiouL06RJkxxYHQAgqI1j2wGAO/LasH3DDTe4ugQAQD26R1euOlLfVJKQkMp2AOCpvDZsO1J2draSk5N19uxZtWvXTr1791a7du1cXRYAeDSz2aTJkypXHanLIxNNrLcNwKMRtm2wevVqrV692vqzr6+vEhMTNW3aNJnN5npfGxAQIB8f7130JTAw0NUlwAPQT1CXkb+XWrQ4q7nPlSgvr8K6/bJ2Ppo+zV+3DLzEhdXBXXFNga3coa8QtuvRpk0b/b//9/900003KTQ0VL/++qu2b9+uF154QcnJyTKZTJo+fXq9xygsLGykahtfYGCgCgoKXF0G3Bz9BBdyXay04j1DP/zYWtmHiv/vCZKGzOYSFRSUuLo8uBmuKbBVY/QVW8I8YbseV199ta6++mrrz/7+/ho4cKC6d++uESNG6J133tG9996roKAgF1YJAJ7PbDYprqefrr6KKSMAvIv3zm9wouDgYN18880qLy/Xzp07XV0OAAAA3BRh205VXxv8+uuvLq4EAAAA7oqwbaeqEe3Q0FAXVwIAAAB3RdiWdPLkSR08eFAnT1Z/TNmePXtqbb9s2TJt2bJF4eHh6tatW2OUCAAAAA/ktTdIpqSkaNu2bZKkAwcOWLelp6dLkmJjYzV69GhJ0rvvvqslS5Zo4sSJ1Z4U+fDDD8vX11ddu3ZVu3bt9Ouvv2rnzp3au3evWrdurfnz519w6T8AAAA0XV4btrdt26ZVq1ZV25aZmanMzEzrz1Vhuy533nmnNmzYoK1bt+rUqVPy8fFR+/btNW7cOI0fP16XXXaZU2oHAACAdzAZhlH3o7vQYN68FihrncIW9BPYir4CW9BPYCt3WWebOdsAAACAkxC2AQAAACchbAMAAABOQtgGAAAAnISwDQAAADgJYRsAAABwEsI2AAAA4CSEbQAAAMBJCNsAAACAkxC2AQAAACchbAMAAABOQtgGAAAAnISwDQAAADgJYRsAAABwEsI2AAAA4CSEbQAAAMBJCNsAAACAkxC2AQAAACfxdXUBAAAAtbFYDO3cJZ04KQW1kbpHu7oi4OIRtgEAgNtZt97QosWG8vN/2xYcLD0x/ayui3VdXcDFYhoJAABwK+vWG3piVvWgLUn5+dKUqae1br3hmsIAOxC2AQCA27BYKke06/PiEkMWC4EbnoGwDQAA3MbOXaoxon0uw5COH69sB3gCwjYAAHAbJ046th3gaoRtAADgNoLaOLYd4GqEbQAA4Da6R1euOlIXk0kKCWEZQHgOwjYAAHAbZrNJkyeZ6m3zyESTzOb62wDugrANAADcSv9+Jj0z21RjhDskRFr4fEv170fQhufgoTYAAMDt9O9nUnwf1XiCZNu2l6igoMTV5QE2I2wDAAC3ZDabFHOtq6sAGoZpJAAAAICTELYBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAADgJIRtAAAAwEkI2wAAAICTELYBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAADgJL6uLsBZPv74Y23btk179uzRgQMHVFZWprlz5+r222+/qONUVFTo3Xff1YoVK3To0CH5+/vrhhtu0JQpU9SxY0cnVQ8AAABv0KCwXVpaqh9++EEnT55UUVGRWrdurTZt2ujKK69Us2bNHFWjXV588UXl5uYqMDBQISEhys3Ntes4s2bNUkpKiq6++mrdfffdOn78uD799FN9++23ev/99xUeHu7YwgEAAOA1Ljpsnzx5Uh9++KHWrVunXbt2qbS0tEabZs2aKTo6WjfeeKNuu+02tWnTxiHFXoynn35aYWFhCg0N1RtvvKEXXnjhoo+xefNmpaSkqGfPnvrHP/5h/QAxfPhw3XfffZozZ47+/ve/O7p0AAAAeAmbw/ahQ4f04osv6rPPPlNZWZkkKTAwUF26dFFAQIBatmyp4uJiFRUV6YcfftDWrVu1detWLVq0SIMGDdLDDz+ssLAwp72R891www0NPkZKSook6ZFHHqk2Ut+/f3/FxcVpw4YNOnLkiNq3b9/gc8H7WCyGdu6STpyUgtpI3aMls9nk6rIAAEAjsilsz549WytWrFBFRYV69eql4cOHKy4urt45y4cPH9bmzZu1evVqffrpp0pLS9Mdd9yhmTNnOqx4Z9uyZYv8/f0VExNTY1/fvn2Vnp6u9PR0jRw5svGLg1tbt97QosWG8vN/2xYcLE2eJPXvR+AGAKCpsClsr1y5UnfddZcmTJigdu3a2XTgjh07qmPHjho9erTy8vK0dOlSffDBBx4TtktKSpSfn6+IiAiZzeYa+6tG6Q8dOtTYpcHNrVtv6IlZRo3t+fnSE7MMPTObwA0AQFNhU9j+/PPPFRwcbPdJ2rVrpyeffFL333+/3cdobMXFxZKkli1b1rq/antVu7oEBATIx8d7V1gMDAx0dQluxWIx9NLLpyTVDNuSZDJJi182KWH4pU1qSgn9BLair8AW9BPYyh36ik1huyFB2xnH8SSFhYWuLsFpAgMDVVBQ4Ooy3ErmdkN5ebUHbUkyDOlYXoW+XlegmGubRtimn8BW9BXYgn4CWzVGX7ElzHvvkGsDtWrVSpJ0+vTpWvdXba9qB0iVN0M6sh0AAPBsNo1sHzlypEEn8cTVOvz9/RUcHKyffvpJFoulxrztqrnajbnCCtxfkI2rXNraDgAAeDabwvaAAQNkMtn3lbfJZNLevXvteq2rxcXF6d///rcyMzPVs2fPavu++eYbSaqxHU1b9+jKVUfOXYXkfCEhle0AAID3syls1xYoy8rKtGPHDkmVNwFWjV4fOXJEhYWFMplM6t69u/z8/BxXrZOcPHlSBQUFCgwMrPYAnjFjxujf//63XnzxxWoPtVm3bp3S09MVHx+v0NBQV5UNN2Q2mzR5kmpdjaTKIxNNTermSAAXx2KxaNOmTcrLy1O7du3Uu3fvWlfFAuAZTIZh1J0K6nD69GmNGzdOpaWlmjZtmvr27Vtt/4YNGzR//nz5+vpq2bJlda7o4UwpKSnatm2bJOnAgQP67rvvFBMTY532ERsbq9GjR0uSFi9erCVLlmjixImaNGlSteM8+eST1se19+/fX/n5+frkk0/UokUL/etf/9IVV1xRbx3efBMHN6nUrbZ1tkNCKoN2U1v2j34CW9FXpNTUVM2cOVM5OTnWbZ06ddKcOXOUkJDgwsrcB/0EtnKXGyQv+nHtkvTiiy/q8OHDWrNmTa2PYo+Pj1dUVJSGDBmiRYsW6cknn7TnNA2ybds2rVq1qtq2zMxMZWZmWn+uCtv1mT17tiIiIrRixQq9/fbb8vf31y233KIpU6aoU6dODq8b3qF/P5Pi+4gnSAKwWWpqqpKSkjR48GAtXbpUnTt3VlZWlhYuXKikpCQlJycTuAEPZNfI9o033qjo6Gi99NJL9bZ7+OGHtXPnTq1bt87uAj2dN3/6ZnQBtqCfwFZNua9YLBbFxsYqKipKy5cvr/Z8hoqKCiUmJiorK0sZGRlNfkpJU+4nuDgePbJ98uRJlZeXX7BdeXk5fyEA1Co7O9ur16HHxWvdurWKiopcXYZL7NixQzk5OZo2bZp2795dY/+wYcO0Zs0aLV++XD169Gj8At1IU+4nqF9AQIDCw8NdXUYNdoXt0NBQbdq0SUePHtXll19ea5ujR49q06ZN3EAIoIYTJ07ouuuuU0VFhatLAdzKxIkT690/ZcqURqoE8Dxms1n79u1TUFCQq0upxq6wPWrUKM2fP1+JiYmaNGmSbr31Vl1yySWSpNLSUn3yySdavHixzpw5o1GjRjm0YACeLygoSBkZGYxso5qmPGK5Y8cOTZkyRUuWLFGXLl1q7N+zZ48mTZqkhQsXMrLdhPsJ6hcQEOB2QVuyc852RUWFpk6dqk8++cS6/nbVjZInT1Y+Gs8wDA0ZMkQLFiyoNvesqfHmaTTMm4Mt6CewVVPuK8zZtl1T7ie4OO4yZ9uuFOzj46MFCxZowYIFio2Nla+vr06cOKETJ07I19dXsbGxeuGFF7Ro0aImHbQBALCF2WzWnDlzlJaWpsTERKWnp6u4uFjp6elKTExUWlqaZs+e3eSDNuCJ7BrZPl95eblOnTolSbr00kvl62vX7BSv5M2fvhldgC3oJ7AVfaX2dbbDwsI0e/Zslv37P/QT2MpdRrbtSsVbt25V27ZtrQ908fX1Vdu2bWu0y87OVn5+Po80BwDABgkJCRo6dChPkAS8iF1h++6779btt9+uv/3tb/W2e/PNN7Vy5UplZWXZVRwAAE2N2WxWfHy8q8sA4CB2T6i2ZfaJA2aoAAAAAB7LqXcvHj9+XP7+/s48BQAAAOC2bJ5G8tFHH1X7OScnp8a2KuXl5frxxx+1adMmde/evSH1AQAAAB7L5rA9ffp065raJpNJmZmZyszMrLO9YRi65JJL9NBDDzW8SgAAAMAD2Ry2H3roIZlMJhmGoZdfflmdO3fWzTffXGtbPz8/hYSEqE+fPgoJCXFYsQAAAIAnsTlsT5o0yfrfq1atUu/evTVx4kSnFAUAAAB4A7uW/vvyyy8dXQcAAADgdRz6qMfy8nKlpKTo+++/12WXXaY77rhDAQEBjjwFAAAA4DHsCttLlizRyy+/rLffftv6dMiKigrdfffd2rFjhwzDkMlkUkpKilauXKnWrVs7tGgAAADAE9i1zvbGjRt12WWXVXsM+5o1a7R9+3ZFRERo9uzZuvnmm3X48GG9++67DisWAAAA8CR2he3Dhw/ryiuvrLbts88+k8lk0oIFCzRmzBgtXrxYl19+udLS0hxSKAAAAOBp7Arbp06dUmBgYLVtW7duVVhYmH73u99JqlyLu1u3bjp69GjDqwQAAAA8kF1hOzAwUHl5edaf//Of/+jnn39WXFxctXZ+fn4qLS1tWIUAAACAh7IrbP/ud7/T9u3btXfvXknSW2+9JZPJpP79+1drl5uby0NtAAAA0GTZtRpJUlKSNm3apFGjRqlVq1YqKipShw4d1LdvX2ub4uJi7dmzRwMGDHBYsQAAAIAnsWtku3///po5c6batWuns2fPKiYmRkuWLFGzZs2sbT766COVl5erd+/eDisWAAAA8CQmwzAMZxz4zJkzKisrk7+/v8xmszNO4REKCgpcXYLTBAYGevX7g2PQT2Ar+gpsQT+BrRqjr5y/YEhtHPoEyXM1b95czZs3d9bhAQAAALdn1zQSAAAAABdmU9geM2aMvvnmmwadaN26dRo9enSDjgEAAAB4EpumkRQWFuq+++5TZGSkRo4cqaFDh9q0pF9eXp7+/e9/6+OPP9b+/ft1xRVXNLhgAAAAwFPYdIOkxWLRv/71L73yyis6ceKEfHx81KlTJ3Xr1k1XXHGFAgIC1KJFC/3yyy86deqUfvzxR+3evVs5OTkyDENt27bVgw8+qDvuuKPJ3SzpzTdxcJMKbEE/ga3oK7AF/QS28qgbJM1ms/74xz9q9OjRWrNmjVJSUpSZmans7GxJlY9mr1KV3c1ms3r27KkxY8Zo0KBB1ZYFBAAAAJqCi1qNpFmzZhoxYoRGjBih06dPa/v27dq/f79OnDih06dPq2XLlgoKCtI111yja6+9Vi1atHBW3QAAAIDbs3vpv5YtW6pv377VnhoJAAAA4Dcs/QcAAAA4CWEbAAAAcBLCNgAAAOAkhG0AAADASQjbAAAAgJMQtgEAAAAnIWwDAAAATkLYBgAAAJzEpofaHDlypEEnad++fYNeDwCAvSwWizZt2qS8vDy1a9dOvXv3ltlsdnVZAJoIm8L2gAEDZDKZ7DqByWTS3r177XotAAANkZqaqpkzZyonJ8e6rVOnTpozZ44SEhJcWBmApsKmsN2zZ88a28rKyrRjxw5JUkBAgHX0+siRIyosLJTJZFL37t3l5+fnuGoBALBRamqqkpKSNHjwYC1dulSdO3dWVlaWFi5cqKSkJCUnJxO4ATidyTAM42JfdPr0aY0bN06lpaWaNm2a+vbtW23/hg0bNH/+fPn6+mrZsmVq2bKlwwr2NAUFBa4uwWkCAwO9+v3BMegnsJUj+4rFYlFsbKyioqK0fPly+fj8dotSRUWFEhMTlZWVpYyMDKaUeBiuKbBVY/SVwMDAC7axaWT7fC+++KIOHz6sNWvWqE2bNjX2x8fHKyoqSkOGDNGiRYv05JNP2nMaAJKys7NVWFjo6jLs1rp1axUVFbm6DHgAR/aVHTt2KCcnR9OmTdPu3btr7B82bJjWrFmj5cuXq0ePHg45JxoH1xTHCAgIUHh4uKvLaBLsGtm+8cYbFR0drZdeeqnedg8//LB27typdevW2V1gQ+zatUuLFy/W9u3bVV5eroiICCUlJWno0KE2vf7DDz/UjBkz6tz/9ttvq1evXvUew5s/fTO64HwnTpxQZGSkKioqXF0KAMCLmM1m7du3T0FBQa4uxWk8emT75MmTKi8vv2C78vJyl4WxzZs3a8KECWrWrJmGDRumFi1aaO3atZoyZYqOHTum8ePH23ysm2++WZ07d66xPTQ01JElAzUEBQUpIyODkW00CY4e2Z4yZYqWLFmiLl261Ni/Z88eTZo0SQsXLmRk28NwTXGMgIAArw7a7sSusB0aGqpNmzbp6NGjuvzyy2ttc/ToUW3atMklgbS8vFwzZ86UyWTSu+++aw3KDz30kEaNGqUFCxZo8ODBNtc2cOBA3X777c4sGaiTp3/NxzcgsJUj+0rXrl21cOFCrV69WnfeeWeNOdvPPvuswsLClJiYyJxtD8M1BZ7GrofajBo1Sr/++qsSExP10Ucf6ezZs9Z9paWl+uijj5SYmKgzZ85o1KhRDivWVps3b1ZOTo6GDx9ebUS6VatWeuCBB1RWVqZVq1Y1el0AgMZhNps1Z84cpaWlKTExUenp6SouLlZ6eroSExOVlpam2bNnW4O2xWLRhg0btHLlSm3YsEEWi8XF7wCAt7BrZPuee+7Rd999p08++UQzZszQjBkzrDdKnjx5UpJkGIaGDBmie+65x3HV2ig9PV1S5Y2a56vatnXrVpuPt3fvXp06dUrl5eXq0KGDevfubdMcHQCA6yQkJCg5OVkzZ87UkCFDrNvDwsKqLfvHWtwAnMmusO3j46MFCxZo4MCBeu+997Rz506dOHFCkuTn56fu3bvrrrvu0rBhwxxarK2ys7MlVV5QzxccHCx/f38dOnTI5uO988471X5u3ry5HnroId13330NqhMA4FwJCQkaOnRonU+QZC1uAM5m12ok5ysvL9epU6ckSZdeeql8fe3K8A4zfvx4ffvtt1q7dm2tgbtv374qKSnRtm3b6j1Oenq6Dhw4oPj4eF122WUqLCzUpk2btGDBAuXl5enJJ5/U3XffXe8xKioqqs0VBAC4B4vFoquuukrdunXTRx99VGNe98iRI7Vnzx59//33zOsGYDeHpGJfX1+1bdvWEYdyK3FxcYqLi7P+3Lx5c40cOVJdunTRH/7wBy1ZskR33XVXvR8uPHkViQvhJhXv5qj1vVk5ALZq7L6yY8cOZWdna+rUqVq/fn2N/YMGDVJqaqoWL17MiiVuhGuKa3jiutwevfSfu6t6YmVxcXGt+0+fPq2AgAC7j3/11VcrNjZWGzdu1MGDBxUZGWn3sQB3dOLECV133XWs740mYeLEifXunzJlSiNVArivprAut7PYFLbHjh1r9wlMJpOWLVtm9+vtUfXJ69ChQ+ratWu1ffn5+SopKVF0dHSDzlH1SebXX39t0HEAd+TI9b0ZhYKtXDGyzVrcnodrimuwLrf9bArbVat72MNkMtn9Wnv17NlTr7/+ujZs2FDjJs0NGzZY29jLYrFoz549kqT27dvbXyjQCCwWQzt3SSdOSkFtpO7Rktl84b+Xjvq6kOlG3sVisdR5s2FDNXZfYS1uz+Rp1xR7r8HwHjaF7S+++MLZdThU79691bFjR61evVpjx461rrVdXFys1157TX5+fho5cqS1/fHjx1VcXKyQkBC1atXKun3Pnj01RsYtFouef/55HTp0SL169VJISEijvCfAHuvWG1q02FB+/m/bgoOlyZOk/v242OPieNsSeVVrcSclJSkxMVGTJ0+2rkayaNEipaWlKTk5maANu3ENhuSg1UjcUV2Pa8/NzdXjjz9e7XHt06dP16pVqzR37txqT4qMjIy0/q9du3YqLCxUenq6srOzddlll2n58uXq2LFjvXV40qfvi+VpowtNzbr1hp6YVfdf72dmmxrlYk8/8Q7nLpE3ZcqUakvkVYXShgZuV/WV2j5EhIWFafbs2R75IcLbeco1xV2uwU2Zu9wg6bVhW5J27dqll156Sdu3b1d5ebkiIiJ0zz33aOjQodXa1RW2n332We3YsUM5OTkqLCyUn5+fOnXqpJtuukn33HOPTTdZesIFwV6ecsFriiwWQ6PurD6acr6QECnlnyanf51JP/F8FotFsbGxioqK0vLly2tMt0hMTFRWVpYyMjIaNArsyr7izOkxcCxPuKa40zW4KSNsNxHufkFoCE+44Dmao5bDc7Z9+ww998KF/2pP+38mXXONcy/03Mzk+RrrRkL6CmzhCf3k3Guw2RygZpeE19rupYUmxVxL2HYWjwrbN998s0wmk9566y117NhRN998s81FmEwmff755za39zbeHEabWtg+ceKEIiMjWQ4PAHARzIqI2itf35orefzPTJNuuZmw7SzuErZtukEyNzdXUuWTIs/9GWhKHLkcnrMxsg1HYmQb7sQT+sn5I9u1BW2pcnUSeD+mkTiZN4/8NrWRbU/iTvMF6SeerynM2Ybn8IR+4k7X4KbMXUa2fS7YAoDHMZtNmjyp/gv4IxO5yMM2VUvkpaWlKTExUenp6SouLlZ6eroSExOVlpam2bNnc0Mh8H+4BuNchG3AS/XvZ9Izs00KDq6+PSSEJadw8RISEpScnKy9e/dqyJAhCgsL05AhQ5SVleWQZf8Ab8M1GFUaPI3kyJEjys/PV2lpaZ1tGvK0Rk/n7l91NYQnfJUH1z+9jH7iXbzpCZLwTJ7WT1x9DW7K3GUaiU03SNYmJSVFr732mo4cOXLBtllZWfaeBkADmc0mxVzr6irgLcxms+Lj411dBuAxuAbDrrD97rvv6umnn5ZhGIqKilKHDh3UokULR9cGAAAAeDS7wvayZcvk5+en119/Xb1793Z0TQAAAIBXsOsGyWPHjqlXr14EbQAAAKAedoXtyy+/XM2bN3d0LQAAAIBXsSts33bbbdqyZYtOnTrl4HIAAAAA72FX2L733nt13XXXaezYsdqyZYt4CCUAAABQk103SJrNZs2ePVtJSUlKSkqSr6+vgoODZTLVXDfSZDLp888/b3ChAAAAgKexK2wfPHhQY8eO1cmTJ2UYhsrKymxabxsAAABoSuwK2/PmzdOJEyd02223KSkpSR07dpS/v7+jawMAAAA8ml1hOzMzU5GRkZo7d66j6wEAAAC8hl03SPr5+emqq65ydC0AAACAV7FrZLtXr17av3+/o2sBAMCrWCyGdu6STpyUgtpI3aMls7nmYgIAvJddYXvq1KkaNWqUFi5cqIcfflhms9nRdQEA4NHWrTe0aLGh/PzftgUHS5MnSf37EbiBpsKusP3xxx/rpptu0htvvKFPP/1UcXFxateuXZ1L/z300EMNLhQAAE+xbr2hJ2bVfAZFfr70xCxDz8wmcANNhcmw44k011xzjUwmk00PszGZTMrKyrKrOG9QUFDg6hKcJjAw0KvfHxyDfgJbeUtfsVgMjbqz+oj2+UJCpJR/mphSYgdv6SdwvsboK4GBgRdsY9fINquQAABQu527VG/QlqTjxyvbxVzbODUBcB27wvZtt93m6DoAAPAKJ046th0Az2bX0n8AAKB2QW0c2w6AZyNsAwDgQN2jK1cdqU9ISGU7AN6PsA0AgAOZzSZNnlT/jY+PTOTmSKCpIGwDAOBg/fuZ9MxsU40R7pAQ6ZnZJpb9A5oQu26QBAAA9evfz6T4PuIJkkATR9gGAMBJzGYTy/sBTRzTSAAAAAAnIWwDAAAATmLTNJIjR4406CTt27dv0OsBAAAAT2RT2B4wYIBMJvtu6DCZTNq7d69drwUAAAA8mU1hu2fPns6uAwAAAPA6NoXtd955x9l1AAAAAF6HGyQBAAAAJyFsAwAAAE7S4IfafP/99zp06JBOnz5dZ5uRI0c29DQAAACAx7E7bG/cuFFPPfWUcnJy6mxjGIZMJhNhGwAAAE2SXWF79+7duu+++2QymTR8+HAdOHBABw4c0H333aecnBxt3LhRRUVFuv3221ljGwAAAE2WXWH7jTfekMVi0Ztvvqk+ffpoxowZOnDggKZMmSJJKioq0qxZs/T1119r5cqVDi0YAAAA8BR23SC5fft2de7cWX369Kl1f+vWrfXcc8/Jx8dHixYtakh9AAAAgMeyK2yfOnVK4eHh1p/9/PwkSSUlJdZtzZo1U2xsrL799tuGVQgAAAB4KLvCdps2baqtPhIYGChJOnz4cLV2Z8+erXeVEgAAAMCb2RW2w8LC9NNPP1l/jo6OlmEYev/9963bDh06pM2bN6tjx44NrxIAAADwQHbdINmvXz8tWLBABw8e1O9+9zv17dtX7du31z//+U/t3r1bl112mTZv3qyzZ8/qD3/4g6NrttmuXbu0ePFibd++XeXl5YqIiFBSUpKGDh1q8zFKS0v1xhtv6H//93919OhRBQQE6KabbtLkyZMVFBTkxOoBAADg6ewK2yNHjlSrVq1UUVEhqXJ+9quvvqrJkydr9+7d2r17t3x8fDR69GiNGzfOoQXbavPmzZowYYKaNWumYcOGqUWLFlq7dq2mTJmiY8eOafz48Rc8RkVFhf785z9rw4YN6tGjhwYNGqRDhw4pJSVFmzZt0ooVK9SmTZtGeDcAAADwRCbDMAxHHvDgwYMqKipSWFiYy4JoeXm5br31Vh07dkwrVqxQ586dJUnFxcUaNWqUcnNzlZaWptDQ0HqPs3LlSv3lL3/R8OHD9fzzz8tkMkmS/vnPf+qvf/2r7rjjDs2ePbveYxQUFDjmTbmhwMBAr35/cAz6CWxFX4Et6CewVWP0lar7Futj15zt+vzud7/Ttdde69IR382bNysnJ0fDhw+3Bm1JatWqlR544AGVlZVp1apVFzxOSkqKJOnRRx+1Bm1JuvPOO9WxY0elpqbqzJkzjn8DAAAA8Ap2he1ffvlF+/bt08mTJ+tsc/LkSe3bt6/acoCNJT09XZIUHx9fY1/Vtq1bt9Z7jLNnz2rnzp264ooraoyAm0wm3XDDDSopKdGePXscVDUAAAC8jV1h+6233tJtt91WY6m/cx0+fFi33Xab3n77bbuLs1d2drakylVTzhccHCx/f38dOnSo3mPk5OSooqKi2nri56raXnUuAAAA4Hx23SD51VdfqVOnTurevXudbbp3765OnTrp888/1wMPPGB3gfaoWtu7VatWte5v2bKliouL6z1G1f6WLVvWeYxzz1WXgIAA+fg4fLaO27BlrhJAP4Gt6CuwBf0EtnKHvmJX2P7pp58UExNzwXZXXnmltm/fbs8pvEZhYaGrS3AablKBLegnsBV9Bbagn8BWHn2D5JkzZ9S8efMLtmvevLlL5mxXjTrXNXp9+vTpOke9q1Ttr2vkump7XSPfAAAAgF1h+7LLLtPu3bsv2G737t0KCQmx5xQNUjWfurZ52fn5+SopKal1Pve5OnbsKB8fnzrnZFdtr2tONwAAAGBX2O7bt69yc3OVnJxcZ5tly5bpp59+Ut++fe2tzW49e/aUJG3YsKHGvqptVW3q0rx5c0VHR+vHH39Ubm5utX2GYWjjxo3y9/dX165dHVQ1AAAAvI1dYXvChAlq2bKlnn32Wd1///364osv9P333+v777/XF198ofvvv1/z5s1Ty5YtNWHCBEfXfEG9e/dWx44dtXr1amVlZVm3FxcX67XXXpOfn59Gjhxp3X78+HEdPHiwxrSTMWPGSJIWLFigc5/9869//UuHDx9WQkKCTdNpAAAA0DTZ/QTJjIwMTZo0SQUFBdUe+CJVjvwGBgbqxRdfVFxcnEMKvVh1Pa49NzdXjz/+eLXHtU+fPl2rVq3S3Llzdfvtt1u3V1RU6N5777U+rr1nz57KycnR2rVrFRoaqpSUlAs+vMebb+LgJhXYgn4CW9FXYAv6CWzlLjdI2rUaiSRdd911WrNmjVasWKHNmzfr6NGjkqTLL79cvXv31ujRoxUQEGDv4Rvs+uuv13vvvaeXXnpJn3zyicrLyxUREaGpU6dq6NChNh3Dx8dHr776qt544w19/PHHSk5O1qWXXqpRo0Zp8uTJLn1KJgAAANyf3SPbsI03f/pmdAG2oJ/AVvQV2IJ+Alu5y8i2XXO2t2zZYnPbJUuW2HMKAAAAwOPZFbYffPBBfffddxdst2jRIr388sv2nAIAAADweHY/R/y+++6rcw1qSVq4cKFee+011qEGANTJYjGUud3QZ18YSt9aJouFmY0AvItdYfuVV17R6dOnNX78eOXl5dXY//zzz+v111/XlVdeqbfffrvBRQIAvM+69YZG3Wno4SmGnppj6J4JRRp1p6F16wncALyHXWG7V69eeuGFF5SXl6fx48dXm3z+7LPP6s0339RVV12lt99+W8HBwQ4rFgDgHdatN/TELEP5+dW35+dLT8wicAPwHnZPIxk4cKDmzJmjH374Qffee69++eUXzZ07V2+99ZauvvpqLVu2TG3btnVkrQAAL2CxGFq0uP4w/eISgyklALyC3etsS9Ltt9+uwsJCPfvssxoyZIjy8/OtQZs1qAEAtdm5SzVGtM93/Hhlu5hrG6cmAHCWBoVtSbrnnnt06tQpvf7667rmmmv01ltv2bTmIACgaTpx0rHtAMCd2RS2x44de+ED+VYe6pFHHqm23WQyadmyZXaUBgDwRkE2fvFpazsAcGc2he309HSbDrZv374a20wm08VVBADwat2jpeDg+qeShIRUtgMAT2dT2P7iiy+cXQcAoIkwm02aPKly1ZG6PDLRJLOZwRoAns+msB0aGursOgAATUj/fiY9M1tatLj68n8hIZVBu38/gjYA79DgGyQBALBH/34mxfepXHXkxEkpPKyVrryimBFtAF6FsA0AcBmz2WRd3i8w0E8FBQRtAN7F7ofaAAAAAKgfYRsAAABwEsI2AAAA4CTM2QaAJsZisWjTpk3Ky8tTu3bt1Lt3b5nNZleXBQBeibANAE1IamqqZs6cqZycHOu2Tp06ac6cOUpISHBhZQDgnZhGAgBNRGpqqpKSkhQVFaW0tDTl5OQoLS1NUVFRSkpKUmpqqqtLBACvYzIMo+5HeKHBCgoKXF2C0wQGBnr1+4Nj0E/cg8ViUWxsrKKiorR8+XL5+Pw21lJRUaHExERlZWUpIyPDZVNK6CuwBf0EtmqMvhIYGHjBNkwjAf5Pdna2CgsLXV2G12ndurWKiopcXUaTt2PHDuXk5GjatGnavXt3jf3Dhg3TmjVrtHz5cvXo0aPxCxR9BbZxdD8JCAhQeHi4w44HnI+RbSfz5k/f3jS6cOLECUVGRqqiosLVpQAAGpHZbNa+ffsUFBTk6lLgYIxsA24kKChIGRkZjGw7AaOV7mHHjh2aMmWKlixZoi5dutTYv2fPHk2aNEkLFy5kZBtuzRkj2wRtOBMj207mLSO/tfGmkW04D/3EPTBnG96CfgJbucvINquRAEATYDabNWfOHKWlpSkxMVHp6ekqLi5Wenq6EhMTlZaWptmzZ7PeNgA4GNNIAKCJSEhIUHJysmbOnKkhQ4ZYt4eFhSk5OZl1tgHACQjbANCEJCQkaOjQoTxBEgAaCWEbAJoYs9ms+Ph4V5cBAE0Cc7YBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAADgJIRtAAAAwEkI2wAAAICTELYBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAADgJIRtAAAAwEkI2wAAAICTELYBAAAAJ/F1dQHOcvr0aS1evFhr165Vfn6+QkJCNHjwYE2cOFEtWrSw+TiRkZF17rvttts0b948R5QLAAAAL+SVYbukpESJiYnKyspSfHy8hg0bpqysLP3jH//Q1q1b9e677+qSSy6x+XihoaG67bbbamzv3LmzI8sG4MUsFkM7d0knTkpBbaTu0ZLZbHJ1WQAAJ/PKsP3mm28qKytL9957r6ZOnWrd/vzzz2vp0qVKTk7W/fffb/PxQkNDNWnSJGeUCqAJWLfe0KLFhvLzf9sWHCxNniT170fgBgBv5nVztg3DUEpKivz9/fXggw9W2/fggw/K399fKSkpLqoOQFOzbr2hJ2ZVD9qSlJ8vPTHL0Lr1hmsKAwA0Cq8b2c7Oztbx48cVHx8vf3//avv8/f0VExOjDRs26OjRo7r88sttOmZRUZHef/99FRQUKCAgQDExMfXO5QYAqXLqyKLF9YfpF5cYiu/DlBIA8FZeF7YPHTokSQoPD691f3h4uDZs2KDs7Gybw/a+ffs0a9asatv69u2rZ599VkFBQfW+NiAgQD4+XvcFglVgYKCrS4AHaKr9JH1rmfLzi+ptc/y49MOPrRTX06+RqnJvTbWv4OLQT2Ard+grXhe2i4uLJUktW7asdX/V9tOnT9t0vPHjx2vQoEEKDw+Xn5+fvv/+e73yyitav3697r//fr3//vsym811vr6wsPAi34HnCAwMVEFBgavLgJtryv0k+5BtU0SyDxXr6qsY2W7KfQW2o5/AVo3RV2wJ824btufNm6fS0lKb248dO7bO0eyGePzxx6v9fO211+r111/XuHHjlJ6eri+++EKDBg1y+HkBeL6gNo5tBwDwPG4btt9//32VlJTY3H7w4MEKDw9Xq1atJNU9cl21va6Rb1v4+Pho9OjRSk9PV2ZmJmEbQK26R1euOnL+zZHnCgmpbAcA8E5uG7a3b99u1+vCwsIkVd4oWZuq7Q0dBa/62uBiPhAAaFrMZpMmT6pcdaQuj0w0cXMkgAtirX7P5bZh217h4eEKCQlRZmamSkpKqq1IUlJSoszMTHXo0MHmmyPrsnPnTklShw4dGnQcAN6tfz+TnpmtGutsh4RUBm3W2QZwIazV79m8bpkMk8mk0aNHq6SkRK+88kq1fa+88opKSko0ZsyYatt//fVXHTx4UEeOHKm2ff/+/SorK6txjszMTL355pvy8/PTkCFDHP8mAHiV/v1M+uBfJr200KT/mVn5/yn/JGgDuDDW6vd8XjeyLUkTJkzQF198oaVLlyorK0tRUVHau3evNmzYoG7dumncuHHV2u/atUtjx45VXFyc3nnnHev2t956S19//bViY2N1+eWXy9fXV99//72+/fZbmUwmzZo1S506dWrstwfAA5nNJsVc6+oqAHgS1ur3Dl4Ztv39/bV8+XItXrxYa9eu1ZYtWxQcHKzx48froYceUvPmzW06zs0336yioiLt27dPGzduVFlZmdq2bathw4Zp3Lhxio7mriYAAOAcO3fVf4O1VLlW/85d4sO8GzMZhsH3D07kzWuBstYpbEE/ga3oK7BFU+onn31h6Kk5F45p/zPTpFtuZmT7fO6yzrbXzdkGAADwBqzV7x28choJAABNGcvEeQfW6vcOhG0AALwIy8R5D9bq9w5MIwEAwEuwTJz3qVyr36Tg4OrbQ0KkZ2azhKgnYGQbAAAvwDJx3qt/P5Pi+4ipQR6KsA0AgBdgmTjvxlr9notpJAAAeIETJx3bDoBjELYBAPACLBMHuCfCNgAAXqBqmbj6sEwc0PgI2wAAr2KxWLRhwwatXLlSGzZskMVicXVJjaJymbj6b5hjmTig8RG2AQBeIzU1VbGxsRoxYoTuvfdejRgxQrGxsUpNTXV1aY2CZeIA98NqJAAAr5CamqqkpCQNHjxYS5cuVefOnZWVlaWFCxcqKSlJycnJSkhIcHWZTscycYB7MRmGwQr3TlRQUODqEpwmMDDQq98fHIN+Als1pK9YLBbFxsYqKipKy5cvl4/Pb1/cVlRUKDExUVlZWcrIyJDZbHZUyXABrimwVWP0lcDAwAu2YWQbcGPZ2dkqLCx0dRkN0rp1axUVFbm6DHiAhvSVHTt2KCcnR9OmTdPu3btr7B82bJjWrFmj5cuXq0ePHg2sFK7ENcX5AgICFB4e7uoyvAYj207mzZ++GV1wrhMnTigyMlIVFRWuLgUA0ISYzWbt27dPQUFBri6lQRjZBlCvoKAgZWRkMLKNJqOhI9tTpkzRkiVL1KVLlxr79+zZo0mTJmnhwoWMbHs4rinOFxAQ4PFB250wsu1k3jzyy8g2bEE/ga2Ysw1bcE2BrdxlZJul/wAAHs9sNmvOnDlKS0tTYmKi0tPTVVxcrPT0dCUmJiotLU2zZ88maANodEwjAQB4hYSEBCUnJ2vmzJkaMmSIdXtYWFiTWfYPgPshbAMAvEZCQoKGDh2qTZs2KS8vT+3atVPv3r0Z0QbgMoRtAIBXMZvNio+Pd3UZACCJOdsAAACA0xC2AQAAACchbAMAAABOwpxtAGgkFovFLW7cc5c6AKApIGwDQCNITU3VzJkzlZOTY93WqVMnzZkzp1GXpHOXOgCgqWAaCQA4WWpqqpKSkhQVFaW0tDTl5OQoLS1NUVFRSkpKUmpqapOqAwCaEh7X7mTe/EhZHpkLWzT1fuIujxF3lzrq09T7CmxDP4Gt3OVx7UwjgVvIzs5WYWGhw4535MgRnT592mHHg/1atGihX375xdVluMz+/fuVk5Oju+66SytXrqyxPzo6WmvWrNHcuXMVGRnp9XXUp6n3FdjmsssuU+vWrRv1nAEBAQoPD2/Uc8J7MLLtZN786dtRnxhPnDihyMhIVVRUOKAqAAAcy2w2a9++fQoKCnJ1KbgIjGwD/ycoKEgZGRmMbHuppj5auX//fi1YsECPP/64rrzyyhr7Dx48qOeee06PPvqo00e23aGO+jT1vgLbuGpkm6ANezGy7WSMbKOpa+r9xF3mSrtLHfVp6n0FtqGfwFbuMrLNaiQA4ERms1lz5sxRWlqaEhMTlZ6eruLiYqWnpysxMVFpaWmaPXu20wOuu9QBAE0NI9tO5s2fvhldgC3oJ5VqW986LCxMs2fPdvk6266oozb0FdiCfgJbucvINmHbybz5gsAFD7agn/zGXZ7c6C51nI++AlvQT2Ardwnb3CAJAI3EbDYrPj7e1WW4TR0A0BQwZxsAAABwEsI2AAAA4CSEbQAAAMBJmLMNSZLFYmjnLunESSmojdQ9WjKbTa4uCwDgYfj3BKiOsA2tW29o0WJD+fm/bQsOliZPkvr34wIJALAN/54ANTGNpIlbt97QE7OqXxglKT9femKWoXXrWRkSAHBh/HsC1I6w3YRZLJUjEPV5cYkhi4ULJACgbvx7AtSNsN2E7dylGiMQ5zt+vLIdAAB14d8ToG6E7SbsxEnHtgMANE38ewLUzStvkMzKytKnn36q7777Tt99950KCgoUFxend955x67j7dq1S4sXL9b27dtVXl6uiIgIJSUlaejQoQ6uvHEFtXFsOwBA08S/J0DdvDJsf/7553r99dfl5+enK664QgUFBXYfa/PmzZowYYKaNWumYcOGqUWLFlq7dq2mTJmiY8eOafz48Q6svHF1j668S7y+r/5CQirbAd7OYrFo06ZNysvLU7t27dS7d2+ZzWZqgtN40xJ5/HsC1M0rw/aQIUM0YMAARURE6NSpU4qPj7frOOXl5Zo5c6ZMJpPeffddde7cWZL00EMPadSoUVqwYIEGDx6s0NBQR5bfaMxmkyZPqrxLvC6PTDR57MUfsFVqaqpmzpypnJwc67ZOnTppzpw5SkhIoCY4nLctkce/J0DdvHLO9tVXX60uXbrIz8+vQcfZvHmzcnJyNHz4cGvQlqRWrVrpgQceUFlZmVatWtXQcl2qfz+TnpltUnBw9e0hIdIzs00eedEHLkZqaqqSkpIUFRWltLQ05eTkKC0tTVFRUUpKSlJqaio1waG8dYk8/j0BamcyDMMz/1bbKD8/X/Hx8XbN2V6wYIFef/11LViwQMOGDav1uNdff72WLVtW5zEaMoWlMdnzdWZgYKDHvD+4jjv3E4vFotjYWEVFRWn58uXy8flt/KGiokKJiYnKyspSRkZGo03fcMeaGos79xVHsVgMjbqzZtA+V0iIlPJPzx0Fdvb0mKbQT+AYjdFXAgMDL9jGK6eROEp2drYkKSwsrMa+4OBg+fv769ChQ41c1YVlZ2ersLDwol9n9pFC2lb+9549F27funVrFRUVXfR5anPkyBGdPn3aIceCe2nRooV++eUXV5dRq/379ysnJ0d33XWXVq5cWWN/dHS01qxZo7lz5yoyMrLJ1tRY3LmvOEp2tqHv99c/xnXqpLRggUnh4Z4Zts916qR08D/Vt7Vs2VLt27e3+5j1/dsTEBCg8PBwu48NOAMj2/UYP368vv32W61du7bWwN23b1+VlJRo27ZtdR6joqKi2siUs/38889q166dKioqGu2cAAC4A7PZrGPHjqlt27auLgWwctuR7Xnz5qm0tNTm9mPHjnXLT7P2jDA3hNlsVkZGRqOcl5Ft2MKdRyv379+vBQsW6PHHH9eVV15ZY//Bgwf13HPP6dFHH23UkW13q6mxuHNfcZTsbEPL3rnwGNe4u71jZLs2zh7ZNpvNTDOBJKaRXND777+vkpISm9sPHjzY4WG7ZcuWkqTi4uJa958+fVoBAQEOPacjNNaHDkd24u7duzvkOHA/7jy/0mKx6IMPPtDOnTv12GOP1To/OiwsTDNmzGjUOdvuVlNjcee+4igWi6GN6Rees/3oo547Z9vZmkI/gXdx27C9fft2V5dgDa2HDh1S165dq+3Lz89XSUmJoqNZNBTwVGazWXPmzFFSUpISExM1efJkde7cWVlZWVq0aJHS0tKUnJzskFBr601jjVkTGh9L5AFNj1cu/ecoPXv2lCRt2LChxr6qbVVtAHimhIQEJScna+/evRoyZIjCwsI0ZMgQZWVlKTk52SFrWq9bX7kCxcNTDD01p/L/R91Z9xJvjVETXIcl8oCmhRskJZWVlSknJ0d+fn7q1KmTdXt5ebmGDBmivLw8rVixwrrWdnFxsUaNGqXc3FytWbNGHTp0qPP83vxVF1/lwRae0k+c9bTGqjWV61JfuGpqT5D0lL7iKN70BMnG1NT6CeznLnO2vTJsHzx4UEuXLpUknTlzRp9++qnatm2rvn37WtvMmzfP+t8//fSTbr75ZoWGhurLL7+sdqy6Hteem5urxx9//IKPa/fmCwIXPNiiKfeTprCmsiM15b4C29FPYCt3CdtuO2e7IX7++ecaT3Y8f9u5Ybs+119/vd577z299NJL+uSTT1ReXq6IiAhNnTpVQ4cOdWjdALzLzl2qN2hL0vHjle1irm2cmgAAjcsrw3avXr20f/9+m9t36NCh3vbR0dF68803HVEagCbkxEnHtgMAeB5ukAQAJwlq49h2AADP45Uj2wDch8ViKHO74bE3gTXkJsXu0VJwcP1TSUJCKtsBALwTYRuA06xbb+ill08pL++3+7CDg6XJk+QRy5ulpqZq5syZysnJsW7r1KmT5syZY9Pye6ypDABgGgkAp6ha8i4vr6La9vz8yvBZ1xrT7iI1NVVJSUmKiopSWlqacnJylJaWpqioKCUlJSk1NdWm47CmMgA0bV659J878ebliVh+CXXx9CXvLBaLYmNjFRUVpeXLl9f6yPSsrCxlZGTYPKWENZUvjGsKbEE/ga1Y+g9uITs7W4WFhXa9tnXr1ioqKnJwRfAG+/YZyjlU/+f4Q9nSBx+YdM017hc4d+zYoZycHE2bNk27d++usX/YsGFas2aNli9frh49eth8XLOPFNK28r/37HFQsV6Ea4rnCggIUHh4uKvLANwSI9tO5s6fvk+cOKHIyEhVVFRcuDEAAHUwm83at2+fgoKCnH4uRrZhK0a24XJBQUHKyMhgZBsOt2+foedeuPDn+Gn/z31HtqdMmaIlS5aoS5cuNfbv2bNHkyZN0sKFCy9qZBv145riuQICAholaAOeiJFtJ/PmT9+MLqAuzNmGPbimwBb0E9jKXUa2WY0EgMNVLnlXGaJNdWRpd17yzmw2a86cOUpLS1NiYqLS09NVXFys9PR0JSYmKi0tTbNnzyZoAwAuiLANwCmqlrwLCal+mfGUJe8SEhKUnJysvXv3asiQIQoLC9OQIUOUlZWl5ORkm9bZBgCAaSRO5s1fdfFVHmzRuvWl+npdgccuedeQJ0ji4nBNgS3oJ7CVu0wj4QZJAE5lNpsUc63nhOvzmc1mxcfHu7oMAICHYhoJAAAA4CSEbQAAAMBJCNsAAACAkxC2AQAAACchbAMAAABOQtgGAAAAnISwDQAAADgJYRsAAABwEsI2AAAA4CSEbQAAAMBJCNsAAACAkxC2AQAAACchbAMAAABOYjIMw3B1EQAAAIA3YmQbAAAAcBLCNgAAAOAkhG0AAADASQjbAAAAgJMQtgEAAAAn8XV1AfBsZWVl+vLLL/Xll19q165dOnbsmCTpqquu0m233aY77rhDZrPZxVXCXWRlZenTTz/Vd999p++++04FBQWKi4vTO++84+rS4CK7du3S4sWLtX37dpWXlysiIkJJSUkaOnSoq0uDm/j444+1bds27dmzRwcOHFBZWZnmzp2r22+/3dWlwY3k5eXp008/1fr16/XDDz/o559/VkBAgGJiYjRhwgR1797dZbURttEgOTk5evjhh+Xv76/evXtrwIABKi4u1ldffaWnnnpK69ev16uvviqTyeTqUuEGPv/8c73++uvy8/PTFVdcoYKCAleXBBfavHmzJkyYoGbNmmnYsGFq0aKF1q5dqylTpujYsWMaP368q0uEG3jxxReVm5urwMBAhYSEKDc319UlwQ298847Wrp0qTp16qQ+ffqoTZs2OnTokD7//HN9/vnneuGFF1z2IZ51ttEgeXl5+vzzz3XbbbfJ39/fur2kpER333239uzZo0WLFunWW291YZVwF99//71KS0sVERGhU6dOKT4+npHtJqq8vFy33nqrjh07phUrVqhz586SpOLiYo0aNUq5ublKS0tTaGioiyuFq23cuFFhYWEKDQ3VG2+8oRdeeIGRbdSwdu1aXXrppYqLi6u2PSMjQ0lJSfL399eGDRvUrFmzRq+NOdtokHbt2umPf/xjtaAtSf7+/rrnnnskSVu3bnVFaXBDV199tbp06SI/Pz9XlwIX27x5s3JycjR8+HBr0JakVq1a6YEHHlBZWZlWrVrlwgrhLm644QY+dOGCBg0aVCNoS9J1112nXr16qbCwUPv373dBZYRtOJGvb+UsJeZsAzhfenq6JCk+Pr7GvqptfFAH4AhVeaTq/xsbYRtOs3LlSkm1/2MKoGnLzs6WJIWFhdXYFxwcLH9/fx06dKiRqwLgbY4cOaKNGzcqODhYERERLqmBsA2neP/997V+/Xpdf/316t+/v6vLAeBmTp8+Laly2khtWrZsqeLi4sYsCYCXKSsr07Rp01RaWqqpU6e67Jt2ViOBJGnevHkqLS21uf3YsWMVHh5e676vvvpKc+bMUWhoqObPn++gCuEuHNlXAABwhoqKCk2fPl1bt27VmDFjNHLkSJfVQtiGpMqR6JKSEpvbDx48uNYAtW7dOj388MMKCgrSsmXLFBIS4sAq4Q4c1VfQtLVs2VKS6hy9Pn36tAICAhqzJABeoqKiQn/5y1+0evVqjRgxQk899ZRL6yFsQ5K0ffv2Bh/j66+/1qRJkxQYGKi3335bHTt2dEBlcDeO6CtA1QewQ4cOqWvXrtX25efnq6SkRNHR0S6oDIAnq6io0IwZM/TRRx9p+PDhmjdvnnx8XDtrmjnbcIiqoB0QEKC333671pueAKBKz549JUkbNmyosa9qW1UbALDFuUF76NCheu6559xiRTTCNhps3bp11YI2UwYAXEjv3r3VsWNHrV69WllZWdbtxcXFeu211+Tn5+fSOZYAPEvV1JGPPvpIQ4YM0fz5890iaEs8QRINdPDgQY0cOVKlpaUaNmyYrrjiihptQkNDedIXJFX2l6VLl0qSzpw5o08//VRt27ZV3759rW3mzZvnqvLQyOp6XHtubq4ef/xxHtcOSVJKSoq2bdsmSTpw4IC+++47xcTEWL9BjY2N1ejRo11ZItzA4sWLtWTJEvn7+2vs2LG1rqk9cODAag/RaizM2UaD/Pzzz9aVKf7973/X2iYuLo6wDUmV/eX8pwKev42w3XRcf/31eu+99/TSSy/pk08+UXl5uSIiIjR16lQNHTrU1eXBTWzbtq3GdSMzM1OZmZnWnwnbyM3NlSSVlJTotddeq7VNaGioS8I2I9sAAACAkzBnGwAAAHASwjYAAADgJIRtAAAAwEkI2wAAAICTELYBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAG5g7Nix6tevn0pLS11y/unTpysyMlJbtmyxbjtz5ozi4+N17733uqQmwBsQtgHADSxevFiRkZH68MMPXV0KLtJPP/2kyMhI3X333XYf4+uvv9aWLVt0//33q1mzZnYd48MPP1RkZKQWL15sdx3na968uSZMmKD169dr06ZNDjsu0JQQtgEAcLEFCxaoTZs2Gj16tKtLqeHOO+9U69attWDBAleXAngkwjYAAC60bds27d+/X0OHDrV7VNuZmjdvrkGDBmnXrl3au3evq8sBPA5hG0CjOPer9jNnzuj555/XTTfdpK5du+qWW27RG2+8IcMwan3tqVOn9MILL2jo0KGKjo5WbGysxo4dq6+++qrO823fvl1//vOfdf3116tr164aMGCA/vrXvyovL6/O1+zcuVNTpkxR37591bVrV8XHx2vcuHFasWJFjbZHjx7VrFmzrO+hd+/emjhxonbt2lXvez99+rTmzp2rAQMGqEuXLnrmmWc0YMAALVmyRJI0Y8YMRUZGWv937vzZqhoffvhhxcfHq2vXrurXr5+eeOIJHTlypFq79957T5GRkbrzzjtlsViq7SstLVVCQoIiIyO1evXqOn8f5xowYIAiIyNlGIaWLVumoUOHqlu3burbt6+efvppFRUV1fo6wzC0evVqjR07Vj179lS3bt106623avHixfr1119rtL/77rsVGRmpn376SampqRozZoyuvfZaXXfddTWOec8996hXr17q1q2bBgwYoEceeaTWqQ5Hjx7V7NmzNXDgQHXr1k1xcXG6//77lZmZWaPtli1bFBkZqenTp+vUqVP6n//5H+vvevjw4frggw+qtV+8eLFuvvlmSVJ6enq1P7vp06fb9LutOuawYcNq3Z+ZmakHH3zQ2tf69OmjUaNG6fnnn9cvv/xi/b3NmDFDkrRkyZJqdZw/NemDDz7Q73//e0VHR6tPnz6aPn268vPz661x+PDhkqT333/fpvcE4De+ri4AQNNSVlam8ePH6+DBg4qLi1NJSYm2bt2qF154Qb/88oumTJlSrf2PP/6oe+65R0ePHlVoaKji4+P1yy+/aOfOnXrggQc0bdo0/elPf6r2mo8//lgzZsyQxWJRTEyMLr/8cn333Xf65z//qc8++0xvv/22fve731V7zbJlyzRv3jxVVFSoS5cu6tmzpwoKCrR//34999xzGjNmjLXt/v37NW7cOBUUFOiKK67QoEGDdOTIEX322Wf66quv9Pzzz+vWW2+t8d7PnDmjxMREHTlyRD179lSXLl0UEBCgwYMHa+PGjdq3b59iYmIUFhZmfU3btm2t//3uu+/q6aefliR169ZNsbGx+vHHH/XBBx/oyy+/1PLly63v67//+7+1bt06ff3113r11Vc1ceJE63Hmz5+vAwcOKCEhwRqibDVnzhytWLFCcXFxioiI0NatW/XOO+8oPT1d7733nlq2bGltW1FRoccee0yrV6+Wv7+/unbtqoCAAO3Zs0dLlizR+vXr9c4776h58+Y1zvPGG28oJSVFMTExuummm3T06FFJksVi0aOPPqo1a9bIz89PMTExatu2rY4ePap169aprKxMvXv3th5n+/btuv/++1VYWKgrrrhCN954o06ePKkNGzbom2++0fPPP6+hQ4fWOH9RUZHuuOMOlZSUKDY2VgUFBcrIyNATTzwhwzCs0z06d+6swYMHKy0tTW3btlXfvn2tx4iNjbXpd/r111+refPmio6OrrHvyy+/1EMPPSTDMBQdHa1rr71WRUVFOnTokJYuXao777xTLVq0UN++fVVeXq7MzExdc8016ty5s/UYnTp1sv73888/r6VLl8rPz0+9evVSy5YttX79euuHjLrExMTIz89P69ats+k9ATiHAQCN4PDhw0ZERIQRERFhJCYmGsXFxdZ9u3btMjp37mx0797dOH36tHV7eXm5MXz4cCMiIsJYunSpYbFYrPuys7ONAQMGGJ07dzb2799v3X7kyBEjOjra6Ny5s/H5559bt1ssFuOZZ54xIiIijNtvv71abenp6UZkZKRx7bXXGhs3bqy2r6yszPj666+tP1dUVFhreu6554yKigrrvjVr1hjXXHON0aNHDyMvL6/W937HHXcYhYWFNX4/L730khEREWGsXLmy1t/f9u3bjc6dOxt9+/Y1du/eXW3fihUrjIiICGP06NHVtv/8889G7969jaioKGP79u2GYRjGN998Y0RGRho33XSTUVRUVOu5anPTTTcZERERRkxMTLXznz592hg7dqwRERFhPP3009Ves3TpUuuf9/Hjx63bz549a/zlL38xIiIijPnz51d7TWJiohEREWF069bN2LJlS406Xn75ZSMiIsIYOnSokZOTU21fUVFRtdcUFxcbffr0MTp37mx8/PHH1dru2rXL6Nmzp9GjRw/jxIkT1u2bN2+2/llNmTLFOHv2rHXfZ599ZkRERBg33nhjtWNV/fkmJibW+fury3/+8x9rv6jNH//4RyMiIsJYs2ZNjX07d+6s9vdo5cqVRkREhPHSSy/Veqzt27cbkZGRRmxsrPHdd99Zt5/7ZxgREWFs3ry51tffdtttRkRERI3fO4D6MY0EQKPy8fHRU089VW0EtFu3burXr59+/fVX7dmzx7r9q6++0oEDBzR48GBNmDBBPj6/XbLCwsI0ffp0WSyWatM8UlJSdObMGd16663Wr/erzjt16lSFhIRoz5492rZtm3Vf1RSWBx54oNqoqCT5+vqqf//+1p+3bNmiAwcOqH379po8ebJMJpN13+DBgzVw4ECVlJRo5cqVtb7/J554Qq1bt76YX5m1RovFoqeeekpdu3attm/06NEaMGCAdu7cWW1ObVBQkP72t7+pvLxcjz32mA4fPqzp06fLZDLpueeeU6tWrS66jsTExGrnb9GihWbOnCmTyaQPPvhAZ8+elSSVl5frzTfflL+/vxYuXKjg4GDra5o1a6aZM2cqODhYK1asUEVFRY3zjBo1SnFxcdW2lZaW6q233pIk/e1vf1PHjh2r7W/VqlW113zwwQfKz8/XuHHjNGLEiGptu3XrpgcffFAlJSX63//93xrnb9mypWbOnFltDvXAgQMVERGhI0eO6Keffrrg78oW+/fvlyRdccUVte4/efKkJOmGG26osS86Orra36ML+ec//ynDMDR27FhFRUVZt5/7Z1ifK6+8UpK0b98+m88JgDnbABpZ+/btrf9onys8PFySqs0d3bBhgyTplltuqfVYVV/T796927otIyNDkpSQkFCjfbNmzTRkyBBJsobt8vJypaenS5LuuOOOC9ZfdfwhQ4bIz8+vxv7f//731dqdKzg4WN26dbvgOc5XUVGhTZs26b/+678UHx9fa5uqOc3nzxm/8cYb9d///d/KycnRyJEjlZ+fr3vvvbfaHOiLUduUi6uuukrXXHONSkpKrGF/7969Kigo0LXXXlttKkyV5s2bq0uXLiosLFR2dnaN/QMGDKixbc+ePSoqKtI111yj7t27X7DWb7/9VtKF+09t8+y7dOmiwMDAGttr66cNURWmAwICat3fpUsXSdJjjz2mXbt21frBxFZVfbK2ueFVf4b1ufTSSyX9VjMA2zBnG0Cjuuyyy2rd3qJFC0mq9kCP3NxcSdLUqVM1derUOo9ZUFBg/e/jx49LkkJDQ2ttW7W96kbJU6dO6cyZM7r00kvrDDznqjp+hw4d6j1+VbtztW/f/oLHr01BQYFKSkokqcaodm1tz/f444/riy++UF5eniIjIzVp0iS76pDq/71mZWVZ33fVyO+3335b71zgumq+/PLLa2w7duyYpOpzkOtT1X/uuuuuiz7/xfTThiguLq523PM9+uijOnDggL766it99dVXCggIUExMjG6++WaNGDFCl1xyic3nqvqzqasfVv0Z1qWqxrpuhgVQO8I2gEZ17lSQC6kaxevbt2+to6NVahuBrMuFvipvqPqOfzHB6FxVvwd/f38NHjy43rZXX311jW0ZGRnWoHXs2DGdPHlS7dq1s6sWWxn/t7JMWFiYYmJi6m1bNWJ6Lnt/V+eq+r0NHjxY/v7+dbar7ZuWi+mnDVE1ladqVZHzXX755Vq5cqU2b96sr7/+Wunp6dbg/eabb+pf//rXRfX/hjh9+rQk2TUNCmjKCNsA3FbV6OLo0aMvGDKrhISE6Mcff9SRI0dqDZ5Vo51VYTMwMFDNmzfXqVOnVFRUdMEgERISUu0456sa0a1q5wiBgYG65JJL5OPjo7lz517UB4aCggLNmDFDJpNJw4YN0+rVq/X444/rrbfesuuDR25ubq0j1VVLD1a976rf75VXXql58+Zd9HlqU9UfcnJybG7/448/6r777rvgNwKu0qZNG0lSYWFhnW18fX0VHx9vnUKUm5urv/zlL9q8ebOWLl2qadOm2XSu4OBg5ebm6siRIzVW45FUY/nI81XVWFUzANswZxuA2+rTp48k6bPPPrP5NVVzkWtbP7q0tFRr1qyR9Nt8XbPZbL2pzpY1hKuOv2bNmhrrV0uy3mx3sXOiq+Z/13ZMX19fxcXF6fTp0xf9yOxZs2bp+PHjmjBhgubPn6+4uDht2rTJeqPhxfr0009rbDt48KCysrLk7+9vXXIuOjparVq1Unp6uk6dOmXXuc7XtWtXtW7dWvv27at1nvX5qm4qvJj+Y4+qP7vy8vKLfm3VPOkff/zR5teEhobq3nvvlSR9//33NtdR1Sfr+zOszw8//FCtZgC2IWwDcFuDBg3SVVddpdTUVL388ss15skahqFt27ZVW1lk1KhRat68uT755BN9/fXX1u0VFRVauHCh8vLy1KVLl2prIN97770ymUx67bXXtHnz5mrnKC8vr7a2cK9evRQREaHc3Fy99NJL1R7E89lnn+mzzz6Tv7+//vCHP1zUe60aEa4KNOd74IEH5OPjoxkzZtR40I1UOQ3hgw8+0JkzZ6zbPvjgA61du1ZdunTRww8/LB8fHz377LNq3bq1Fi5caNeqEsuXL6+24smvv/6qp59+WoZh6A9/+IN1zexmzZppwoQJ+uWXXzRp0iQdPny4xrHy8vL00Ucf2XzuZs2aady4cZIqV3U5/9uF4uJi682uUuVjxoOCgvT3v/9d77//fo2bC8vLy/XNN9/owIEDNtdQm8DAQPn5+enw4cO1fliqz5VXXqmgoCBlZWXVGpKTk5NrvRmzqk+eO7e9qg/VFdzvvPNOSZVryp/7Z19SUmL9M6zL2bNndeDAAV1++eU1VoEBUD+mkQBwW76+vnr55Zf1pz/9SS+99JLeffddRUZGqk2bNjp16pSysrJ04sQJzZgxwxqe27dvr6eeekozZszQAw88UO2hNj/++KPatm2r+fPnVztPXFycHnvsMc2fP1/jxo1T165dFR4eroKCAu3bt0+lpaXWlRxMJpOef/55jR07Vq+99po+++wzde7cWUeOHFFmZqZ8fX31zDPPXPQ0kj59+uiSSy7RsmXL9P333yskJEQmk0l/+tOfdOWVV+q6667TrFmzNGfOHI0dO1YREREKDw+Xr6+vcnNzlZWVpdLSUg0aNEjNmzfX4cOH9cwzz6h58+aaP3++ddSzffv2mjVrlqZOnarHHntMH3zwwUXNjx4xYoTGjBmjXr16qVWrVsrIyFB+fr6uvvpqPfLII9Xa3nffffrhhx/08ccf69Zbb1VUVJQ6dOigsrIy/fjjj/rPf/6jyMhIjRw50ubzP/DAA8rKytLnn3+uwYMH67rrrlNQUJCOHj2qvXv36oYbbrB+U9G6dWu98soreuCBBzRr1iy9+uqruvrqq9W6dWv9/PPP2rt3r4qKivTyyy8rIiLC5hrO16xZM8XHx+urr77S73//e0VFRVkfuGPLh67+/fvrww8/1M6dO2s8CGfJkiV69tlndc011ygsLEyGYWjfvn3Kzs7WpZdeqvHjx1vb9ujRQ0FBQUpLS9Pdd9+tDh06yMfHR3/4wx8UExOjmJgYjR8/Xv/4xz80atQo65/h1q1b1axZM9100011PpU1MzNTZWVl1ZbBBGAbwjYAtxYeHq6PPvpIy5cv12effaYdO3bIYrGobdu26ty5swYMGFDjaY0jR45Up06d9MYbb2j79u3atWuXgoODddddd+nPf/5zrTcH/ulPf1L37t2VnJyszMxM7d+/X5deeqkiIiJqLJUWGRmpVatW6dVXX9U333yjtLQ0tWzZUgMHDtT9999f65MAL6Rdu3Z65ZVX9PLLL2vbtm3W1UdGjBhhvYHvrrvuUo8ePbRs2TLrjXL/9V//pXbt2ikhIUGDBg1Sq1atZLFY9Nhjj6mkpESzZs2qMT83ISFBX3/9tVavXq358+frySeftLnOJ598Uh06dFBKSop++uknBQQE6I9//KMeeeSRGut2+/j46LnnntPgwYO1YsUK7d69W3v37lXr1q112WWX6U9/+lOtSwnWx9fXV4sXL9bHH3+slStXas+ePTp79qyCg4N144031lh5pEePHkpNTVVycrLWrVunrVu3Sqqcv9yzZ0/dcsstNdZWt8czzzyjZ599Vhs3btTq1atlsVhksVhsCtujR4/Whx9+qNWrV9cI208++aS++eYbfffdd1q/fr2kytHse+65R/fcc0+1vnzJJZfo9ddf18KFC7Vr1y5t3bpVhmEoNjbWepPq448/riuvvNL61M9WrVqpT58+mjp1qhYuXFhnjampqZJU7UmqAGxjMur73ggAAFWue52bm2t9CAsca+TIkTp27JjWr19f7UE67uDMmTPq27evwsPDlZKS4upyAI/DnG0AAFxsypQpKigoqPY0VHfxr3/9S0VFRXr00UddXQrgkQjbAAC4WP/+/dWrVy+98cYbDntgjiOcOXNGb775pvr16+eQ6TZAU8Q0EgDABTGNBADsQ9gGAAAAnIRpJAAAAICTELYBAAAAJyFsAwAAAE5C2AYAAACchLANAAAAOAlhGwAAAHASwjYAAADgJIRtAAAAwEn+P303/4IMNxkyAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "post = m15_5.get_samples()\n",
    "B_impute_mu = jnp.mean(post[\"B_impute\"], 0)\n",
    "B_impute_ci = jnp.percentile(post[\"B_impute\"], q=jnp.array([5.5, 94.5]), axis=0)\n",
    "\n",
    "# B vs K\n",
    "plt.plot(dat_list[\"B\"], dat_list[\"K\"], \"o\")\n",
    "plt.gca().set(xlabel=\"neocortex percent (std)\", ylabel=\"kcal mild (std)\")\n",
    "miss_idx = pd.isna(dat_list[\"B\"]).nonzero()[0]\n",
    "Ki = dat_list[\"K\"][miss_idx]\n",
    "plt.plot(B_impute_mu, Ki, \"ko\", mfc=\"none\")\n",
    "for i in range(12):\n",
    "    plt.plot(B_impute_ci[:, i], jnp.repeat(Ki[i], 2), \"k\", lw=1)\n",
    "plt.show()\n",
    "\n",
    "# M vs B\n",
    "plt.plot(dat_list[\"M\"], dat_list[\"B\"], \"o\")\n",
    "plt.gca().set(xlabel=\"log body mass (std)\", ylabel=\"neocortex percent (std)\")\n",
    "Mi = dat_list[\"M\"][miss_idx]\n",
    "plt.plot(Mi, B_impute_mu, \"ko\", mfc=\"none\")\n",
    "for i in range(12):\n",
    "    plt.plot(jnp.repeat(Mi[i], 2), B_impute_ci[:, i], \"k\", lw=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.22"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fdc2ce557b134f4aba91004daa1fcd18",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d584f96f31ba41f2bcbcd43e336e0c2e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e83dbee12fda4a6aa0c79e66942812bb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79e4b44eac8a43b68e67f561e2b211c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "                 mean       std    median      5.0%     95.0%     n_eff     r_hat\n",
      "Rho_BM[0,0]      1.00      0.00      1.00      1.00      1.00       nan       nan\n",
      "Rho_BM[0,1]      0.61      0.13      0.62      0.39      0.80   1387.58      1.00\n",
      "Rho_BM[1,0]      0.61      0.13      0.62      0.39      0.80   1387.58      1.00\n",
      "Rho_BM[1,1]      1.00      0.00      1.00      1.00      1.00    297.05      1.00\n",
      "         bB      0.58      0.26      0.60      0.15      0.99    766.01      1.01\n",
      "         bM     -0.64      0.22     -0.65     -1.01     -0.30    860.13      1.00\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def model(B, M, K):\n",
    "    # priors\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 0.5))\n",
    "    muB = numpyro.sample(\"muB\", dist.Normal(0, 0.5))\n",
    "    muM = numpyro.sample(\"muM\", dist.Normal(0, 0.5))\n",
    "    bB = numpyro.sample(\"bB\", dist.Normal(0, 0.5))\n",
    "    bM = numpyro.sample(\"bM\", dist.Normal(0, 0.5))\n",
    "    sigma = numpyro.sample(\"sigma\", dist.Exponential(1))\n",
    "    Rho_BM = numpyro.sample(\"Rho_BM\", dist.LKJ(2, 2))\n",
    "    Sigma_BM = numpyro.sample(\"Sigma_BM\", dist.Exponential(1).expand([2]))\n",
    "\n",
    "    # define B_merge as mix of observed and imputed values\n",
    "    B_impute = numpyro.sample(\n",
    "        \"B_impute\", dist.Normal(0, 1).expand([int(np.isnan(B).sum())]).mask(False)\n",
    "    )\n",
    "    B_merge = jnp.asarray(B).at[np.nonzero(np.isnan(B))[0]].set(B_impute)\n",
    "\n",
    "    # M and B correlation\n",
    "    MB = jnp.stack([M, B_merge], axis=1)\n",
    "    cov = jnp.outer(Sigma_BM, Sigma_BM) * Rho_BM\n",
    "    numpyro.sample(\"MB\", dist.MultivariateNormal(jnp.stack([muM, muB]), cov), obs=MB)\n",
    "\n",
    "    # K as function of B and M\n",
    "    mu = a + bB * B_merge + bM * M\n",
    "    numpyro.sample(\"K\", dist.Normal(mu, sigma), obs=K)\n",
    "\n",
    "\n",
    "m15_7 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_7.run(random.PRNGKey(0), **dat_list)\n",
    "post = m15_7.get_samples(group_by_chain=True)\n",
    "print_summary({k: v for k, v in post.items() if k in [\"bM\", \"bB\", \"Rho_BM\"]})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.23"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "B_missidx = pd.isna(dat_list[\"B\"]).nonzero()[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>polity</th>\n",
       "      <th>year</th>\n",
       "      <th>population</th>\n",
       "      <th>moralizing_gods</th>\n",
       "      <th>writing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Big Island Hawaii</td>\n",
       "      <td>1000</td>\n",
       "      <td>3.729643</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Big Island Hawaii</td>\n",
       "      <td>1100</td>\n",
       "      <td>3.729643</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Big Island Hawaii</td>\n",
       "      <td>1200</td>\n",
       "      <td>3.598340</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Big Island Hawaii</td>\n",
       "      <td>1300</td>\n",
       "      <td>4.026240</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Big Island Hawaii</td>\n",
       "      <td>1400</td>\n",
       "      <td>4.311767</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>859</th>\n",
       "      <td>Yemeni Coastal Plain</td>\n",
       "      <td>1400</td>\n",
       "      <td>6.763083</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>860</th>\n",
       "      <td>Yemeni Coastal Plain</td>\n",
       "      <td>1500</td>\n",
       "      <td>6.519621</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>861</th>\n",
       "      <td>Konya Plain</td>\n",
       "      <td>1600</td>\n",
       "      <td>7.447158</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>862</th>\n",
       "      <td>Yemeni Coastal Plain</td>\n",
       "      <td>1700</td>\n",
       "      <td>3.882606</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>863</th>\n",
       "      <td>Yemeni Coastal Plain</td>\n",
       "      <td>1800</td>\n",
       "      <td>3.882606</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>864 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                   polity  year  population  moralizing_gods  writing\n",
       "0       Big Island Hawaii  1000    3.729643              NaN        0\n",
       "1       Big Island Hawaii  1100    3.729643              NaN        0\n",
       "2       Big Island Hawaii  1200    3.598340              NaN        0\n",
       "3       Big Island Hawaii  1300    4.026240              NaN        0\n",
       "4       Big Island Hawaii  1400    4.311767              NaN        0\n",
       "..                    ...   ...         ...              ...      ...\n",
       "859  Yemeni Coastal Plain  1400    6.763083              1.0        1\n",
       "860  Yemeni Coastal Plain  1500    6.519621              1.0        1\n",
       "861           Konya Plain  1600    7.447158              1.0        1\n",
       "862  Yemeni Coastal Plain  1700    3.882606              1.0        1\n",
       "863  Yemeni Coastal Plain  1800    3.882606              1.0        1\n",
       "\n",
       "[864 rows x 5 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Moralizing_gods = pd.read_csv(\"../data/Moralizing_gods.csv\", sep=\";\")\n",
    "Moralizing_gods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN    528\n",
       "1.0    319\n",
       "0.0     17\n",
       "Name: moralizing_gods, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Moralizing_gods.moralizing_gods.value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.26"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "symbol = Moralizing_gods.moralizing_gods.apply(lambda x: \".\" if x == 1 else \"o\")\n",
    "symbol[Moralizing_gods.moralizing_gods.isna()] = \"x\"\n",
    "color = Moralizing_gods.moralizing_gods.apply(lambda x: \"k\" if pd.isna(x) else \"b\")\n",
    "for pch in [\"o\", \".\", \"x\"]:\n",
    "    plt.scatter(\n",
    "        Moralizing_gods.year[symbol == pch],\n",
    "        Moralizing_gods.population[symbol == pch],\n",
    "        marker=pch,\n",
    "        color=color[symbol == pch],\n",
    "        facecolor=\"none\" if pch == \"o\" else None,\n",
    "        lw=1.5,\n",
    "        alpha=0.7,\n",
    "    )\n",
    "plt.gca().set(xlabel=\"Time (year)\", ylabel=\"Population size\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.27"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>writing</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>moralizing_gods</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>9</td>\n",
       "      <td>310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nan</th>\n",
       "      <td>442</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "writing            0    1\n",
       "moralizing_gods          \n",
       "0.0               16    1\n",
       "1.0                9  310\n",
       "nan              442   86"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dmg = Moralizing_gods\n",
    "dmg.astype(str).groupby([\"moralizing_gods\", \"writing\"]).size().unstack(fill_value=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.28"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <td>1000.00</td>\n",
       "      <td>1100.00</td>\n",
       "      <td>1200.000</td>\n",
       "      <td>1300.000</td>\n",
       "      <td>1400.000</td>\n",
       "      <td>1500.000</td>\n",
       "      <td>1600.000</td>\n",
       "      <td>1700.000</td>\n",
       "      <td>1800.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>population</th>\n",
       "      <td>3.73</td>\n",
       "      <td>3.73</td>\n",
       "      <td>3.598</td>\n",
       "      <td>4.026</td>\n",
       "      <td>4.312</td>\n",
       "      <td>4.205</td>\n",
       "      <td>4.374</td>\n",
       "      <td>5.158</td>\n",
       "      <td>4.997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>writing</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>moralizing_gods</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       0        1         2         3         4         5  \\\n",
       "year             1000.00  1100.00  1200.000  1300.000  1400.000  1500.000   \n",
       "population          3.73     3.73     3.598     4.026     4.312     4.205   \n",
       "writing             0.00     0.00     0.000     0.000     0.000     0.000   \n",
       "moralizing_gods      NaN      NaN       NaN       NaN       NaN       NaN   \n",
       "\n",
       "                        6         7         8  \n",
       "year             1600.000  1700.000  1800.000  \n",
       "population          4.374     5.158     4.997  \n",
       "writing             0.000     0.000     0.000  \n",
       "moralizing_gods       NaN       NaN     1.000  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dmg = Moralizing_gods\n",
    "haw = dmg.polity == \"Big Island Hawaii\"\n",
    "dmg.loc[haw, [\"year\", \"population\", \"writing\", \"moralizing_gods\"]].T.round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.29"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "with numpyro.handlers.seed(rng_seed=9):\n",
    "    N_houses = 100\n",
    "    alpha = 5\n",
    "    beta = -3\n",
    "    k = 0.5\n",
    "    r = 0.2\n",
    "    cat = numpyro.sample(\"cat\", dist.Bernoulli(k).expand([N_houses]))\n",
    "    notes = numpyro.sample(\"notes\", dist.Poisson(alpha + beta * cat))\n",
    "    R_C = numpyro.sample(\"R_C\", dist.Bernoulli(r).expand([N_houses]))\n",
    "    cat_obs = jnp.where(R_C == 1, -9, cat)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.30"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d994f77180249a89c8f7620c3443507",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3c8818cb7da74e45ae285ff17ac781e4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9502cb04aee241828b2a143996398fca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c7a153d546534ac6bae855017dc6fbaf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dat = dict(notes=notes, cat=np.asarray(cat_obs), RC=np.asarray(R_C), N=N_houses - 1)\n",
    "\n",
    "\n",
    "def model(N, RC, cat, notes):\n",
    "    # priors\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 1))\n",
    "    b = numpyro.sample(\"b\", dist.Normal(0, 0.5))\n",
    "\n",
    "    # sneaking cat model\n",
    "    k = numpyro.sample(\"k\", dist.Beta(2, 2))\n",
    "    numpyro.sample(\"cat|RC==0\", dist.Bernoulli(k), obs=cat[RC == 0])\n",
    "\n",
    "    # singing bird model\n",
    "    # cat NA:\n",
    "    custom_logprob = jnp.logaddexp(\n",
    "        jnp.log(k) + dist.Poisson(jnp.exp(a + b)).log_prob(notes[RC == 1]),\n",
    "        jnp.log(1 - k) + dist.Poisson(jnp.exp(a)).log_prob(notes[RC == 1]),\n",
    "    )\n",
    "    numpyro.factor(\"notes|RC==1\", custom_logprob)\n",
    "    # cat known present/absent:\n",
    "    lambda_ = jnp.exp(a + b * cat[RC == 0])\n",
    "    numpyro.sample(\"notes|RC==0\", dist.Poisson(lambda_), obs=notes[RC == 0])\n",
    "\n",
    "\n",
    "m15_8 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_8.run(random.PRNGKey(0), **dat)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.31"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "111aeb5b66424f00a5b4124b790bd267",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c8850f9235954e9781779a5d3c1c2751",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b7b4b9e059e6460ca3b4d94a1052c35e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57ab084fa5eb43bb882f77ccc5f6fd80",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/1000 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def model(N, RC, cat, notes, link=False):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 1))\n",
    "    b = numpyro.sample(\"b\", dist.Normal(0, 0.5))\n",
    "\n",
    "    # sneaking cat model\n",
    "    k = numpyro.sample(\"k\", dist.Beta(2, 2))\n",
    "    numpyro.sample(\"cat|RC==0\", dist.Bernoulli(k), obs=cat[RC == 0])\n",
    "\n",
    "    # singing bird model\n",
    "    custom_logprob = jnp.logaddexp(\n",
    "        jnp.log(k) + dist.Poisson(jnp.exp(a + b)).log_prob(notes[RC == 1]),\n",
    "        jnp.log(1 - k) + dist.Poisson(jnp.exp(a)).log_prob(notes[RC == 1]),\n",
    "    )\n",
    "    numpyro.factor(\"notes|RC==1\", custom_logprob)\n",
    "    lambda_ = jnp.exp(a + b * cat[RC == 0])\n",
    "    numpyro.sample(\"notes|RC==0\", dist.Poisson(lambda_), obs=notes[RC == 0])\n",
    "\n",
    "    if link:\n",
    "        lpC0 = numpyro.deterministic(\n",
    "            \"lpC0\", jnp.log(1 - k) + dist.Poisson(jnp.exp(a)).log_prob(notes)\n",
    "        )\n",
    "        lpC1 = numpyro.deterministic(\n",
    "            \"lpC1\", jnp.log(k) + dist.Poisson(jnp.exp(a + b)).log_prob(notes)\n",
    "        )\n",
    "        numpyro.deterministic(\"PrC1\", jnp.exp(lpC1) / (jnp.exp(lpC1) + jnp.exp(lpC0)))\n",
    "\n",
    "\n",
    "m15_9 = MCMC(NUTS(model), num_warmup=500, num_samples=500, num_chains=4)\n",
    "m15_9.run(random.PRNGKey(0), **dat)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "with numpyro.handlers.seed(rng_seed=100):\n",
    "    x = numpyro.sample(\"x\", dist.Normal().expand([10]))\n",
    "    y = numpyro.sample(\"y\", dist.Normal(x))\n",
    "    x = jnp.concatenate([x, jnp.array([jnp.nan])])\n",
    "    y = jnp.concatenate([y, jnp.array([100])])\n",
    "    d = dict(x=x, y=y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.33"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "Primates301 = pd.read_csv(\"../data/Primates301.csv\", sep=\";\")\n",
    "d = Primates301\n",
    "cc = d.dropna(subset=[\"brain\", \"body\"]).index\n",
    "B = d.brain[cc]\n",
    "M = d.body[cc]\n",
    "B = B.values / max(B)\n",
    "M = M.values / max(M)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.34"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "Bse = B * 0.1\n",
    "Mse = M * 0.1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.35"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "sample: 100%|██████████| 1000/1000 [00:03<00:00, 285.81it/s, 3 steps of size 2.54e-01. acc. prob=0.93]\n"
     ]
    }
   ],
   "source": [
    "dat_list = dict(B=B, M=M)\n",
    "\n",
    "\n",
    "def model(M, B):\n",
    "    a = numpyro.sample(\"a\", dist.Normal(0, 1))\n",
    "    b = numpyro.sample(\"b\", dist.Normal(0, 1))\n",
    "    sigma = numpyro.sample(\"sigma\", dist.Exponential(1))\n",
    "    mu = a + b * jnp.log(M)\n",
    "    numpyro.sample(\"B\", dist.LogNormal(mu, sigma), obs=B)\n",
    "\n",
    "\n",
    "m15H4 = MCMC(NUTS(model), num_warmup=500, num_samples=500)\n",
    "m15H4.run(random.PRNGKey(0), **dat_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.36"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = dict(M_true=dat_list[\"M\"], B_true=dat_list[\"B\"])\n",
    "init_strategy = init_to_value(values=start)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.37"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "name                     0\n",
       "genus                    0\n",
       "species                  0\n",
       "subspecies             267\n",
       "spp_id                   0\n",
       "genus_id                 0\n",
       "social_learning         98\n",
       "research_effort        115\n",
       "brain                  117\n",
       "body                    63\n",
       "group_size             114\n",
       "gestation              161\n",
       "weaning                185\n",
       "longevity              181\n",
       "sex_maturity           194\n",
       "maternal_investment    197\n",
       "dtype: int64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Primates301 = pd.read_csv(\"../data/Primates301.csv\", sep=\";\")\n",
    "d = Primates301\n",
    "d.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.38"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "cc = d.dropna(subset=[\"body\"]).index\n",
    "M = d.body[cc]\n",
    "M = M.values / max(M)\n",
    "B = d.brain[cc]\n",
    "B = B.values / B.max(skipna=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code 15.39"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = dict(B_impute=jnp.repeat(0.5, 56))\n",
    "init_strategy = init_to_value(values=start)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}