Skip to content

Instantly share code, notes, and snippets.

@ColCarroll
Created August 15, 2017 01:06
Show Gist options
  • Save ColCarroll/9fb0e6714dc0369acf6549cededcc875 to your computer and use it in GitHub Desktop.
Save ColCarroll/9fb0e6714dc0369acf6549cededcc875 to your computer and use it in GitHub Desktop.
ecological_inference_edits
Display the source blob
Display the rendered blob
Raw
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Goals\n",
"\n",
"1. Introduce the problem of ecological inference\n",
"\n",
"2. Describe the hierarchical Bayesian approach to 2x2 ecological inference as introduced in [2, 3]\n",
"\n",
"3. Use python and PyMC3 to implement the model and obtain posterior samples using MCMC techniques (with simulated and real data)\n",
"\n",
"4. Provide some familiarity with the use of ecological inference in the context of voting rights [6]\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Contents\n",
"\n",
"The problem\n",
"\n",
"A hierarchical Bayesian Approach\n",
"\n",
"Extending the model to include covariates\n",
"\n",
"Examples with code:\n",
"\n",
"1. A toy example \n",
"\n",
"2. An example with data simulated from the generative model\n",
"\n",
"3. An example with real voting data\n",
"\n",
"4. An example with real voting data and covariates\n",
"\n",
"[References](#References)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# The problem\n",
"In an *ecological inference* problem, broadly speaking, the task is to infer individual-level behavior from the behavior of groups. Does that sounds like a hard problem? It is!\n",
"\n",
"It also has important applications to voting rights.\n",
"\n",
"Let's make the problem more concrete. Here we look at the \"two by two\" case. We have group data about two different binary characteristics of several populations - say, what percentage is falls in some \"Group 1\" and what precentage is in some different and overlapping classification \"Group A\". We would then like to ask questions like \"what percentage of Group 1 members are also in Group A?\". \n",
"\n",
"Put another way, we know the marginal distributions in the following 2 x 2 table, and we want to infer the values of each of the cells marked with question marks.\n",
"\n",
"| Population i | Group A | Not Group A | |\n",
"|----------------------------------------------------------|\n",
"| **Group 1** | *?* | *?* | $X_i $ |\n",
"| **Group 2** | *?* | *?* | $1-X_i $ |\n",
"| | $ T_i $ | $1-T_i $ | |\n",
"\n",
"We imagine having such a table for a number $p$ of different populations: $i = 1,..., p$. We denote the size of population $i$ by $N_i$.\n",
"\n",
"In the area of voting rights, we may be interested in voting rates of different groups. We may also want to know the rates at which voters in one group vote for a particular candidate or party, but this data is unavailable because ballots are secret. \n",
"\n",
"From what I understand, this latter kind of information is actually essential in the area of voting rights, since a racial group being 'politically cohesive' was declared to be one component of a justiciable vote dilution claim in Thornburgh v. Gigles (478 U.S. 30-108) [6][7]\n",
"\n",
"We illustrate with an example from [King 1997] applying ecological inference to $p=268$ precincts in 1968. In the $i$th distict we know the percentage the voting age population that is black ($X_i$) and white ($1-X_i$)†, the proportion of voting age members of the district who are registered to vote ($T_i$), and size of the total voting age population ($N_i$). From this we wish to infer the proportion of black voting age members of the population who are registered to vote and the proportion of white voting age members of the population who are registered to vote.\n",
"\n",
"| District i | Registered | Not Registered | |\n",
"|--------------|--------------|----------------|-----------|\n",
"| **Black** | *?* | *?* | $X_i $ |\n",
"| **White** | *?* | *?* | $1-X_i $ |\n",
"| | $ T_i $ | $1-T_i $ | |\n",
"\n",
"†In the data used in King the population of these 1968 counties is grouped into black and white only.\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# A hierarchical Bayesian approach\n",
"\n",
"Now we describe the hierarchical Bayesian approach to 2x2 ecological inference as introduced in [2, 3]\n",
"\n",
" \n",
"### Observed quantities \n",
"\n",
"$\\begin{align}\n",
"X_i &= \\text{ fraction of population i who are in Group 1} \\\\\n",
"&= \\text{ fraction of voting age people in precinct i who are black}\n",
"\\end{align}$\n",
" \n",
"$\\begin{align} \n",
"T_i &= \\text{ fraction of population i who are in Group A}\\\\\n",
"&= \\text{ fraction of voting age people who are registered to vote in precinct i}\n",
"\\end{align}$\n",
"\n",
"$\\begin{align}\n",
"N_i &= \\text{ size of population i} \\\\\n",
"&= \\text{ number of voting age people in precinct i}\n",
"\\end{align}$\n",
"\n",
"$\\begin{align}\n",
"T'_i &= \\text{ number of people in population i who are in Group A}\\\\\n",
" &= \\text{ number of people in precinct i who are registered to vote}\n",
"\\end{align}$\n",
" \n",
" \n",
"### Unobserved parameters of interest\n",
"$\\begin{align} b1_i &= \\text{fraction of Group 1 members in population i that are in Group A} \\\\\n",
" &= \\text{fraction of black voting age population in precinct i that is registered to vote}\n",
"\\end{align}$\n",
"\n",
"$\\begin{align}\n",
"b2_i &= \\text{ fraction of Group 2 members in population i that are in Group A} \\\\\n",
"&= \\text{ fraction of white voting age population in precinct i that is registered to vote}\n",
"\\end{align}$ \n",
"\n",
"\n",
"### Other unobserved parameters\n",
"$\\begin{align}\n",
"\\theta_i &= X_i * b1_i + (1-X_i) * b2_i \\\\\n",
" &= \\text{ expected fraction of population i that is in Group A} \\\\ \n",
" &= \\text{ expected fraction of the population of precinct i that is registered to vote}\n",
"\\end{align}$\n",
" \n",
"$c_1, d_1$ = hyperparameters for beta distribution that governs each of the $b1_i$\n",
"\n",
"$c_2, d_2$ = hyperparameters for beta distribution that governs each of the $b2_i$\n",
"\n",
"$\\lambda$ = fixed hyperparameter for exponential distributions of $c_1, d_1, c_2, d_2$. In our examples we use $\\lambda = 0.5$ to match [King 1999]\n",
" \n",
" \n",
"### The model\n",
" \n",
" $c_{1} \\sim Exponential(\\lambda)$\n",
" \n",
" $d_{1} \\sim Exponential(\\lambda)$\n",
" \n",
" $c_{2} \\sim Exponential(\\lambda)$\n",
" \n",
" $d_{2} \\sim Exponential (\\lambda)$\n",
" \n",
" $b1_i \\, \\mid \\, c_1, d_1 \\sim Beta(c_1, d_1)$\n",
" \n",
" $b2_i \\, \\mid \\, c_2, d_2 \\sim Beta(c_2, d_2)$\n",
" \n",
" $T'_i \\, \\mid \\, b1_i, b2_i, X_i \\sim Binomial(N_i, \\theta_i)$\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<img src=\"EI_Graphical_Model_5.png\">"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Extending the model to include covariates\n",
"\n",
"We may want our generative model for $b1_i$ and $b2_i$ to take into account additional variables that characterize the the $i$ districts. For instance, we may want the $b1_i$ and $b2_i$ to be drawn from a distribution whose parameters include some spatial information if we suspect, for example, that precincts that are geographically close might show similar patterns. \n",
"\n",
"To that end, suppose we have some covariate $Z_i$ measured for each precinct (here we assume $Z_i$ is a scalar, but the generalization to the case where $Z_i$ is a vector is straightforward). The model can then be extended as follows [King, 1999].\n",
"\n",
"To our list of observed quantities, add $Z_i$.\n",
"\n",
"In our list of unobserved parameters, replace $c_1$ and $c_2$ with $\\alpha, \\beta, \\gamma,$ and $\\delta$. \n",
"\n",
"Give $\\alpha$, $\\beta$, $\\gamma$ and $\\delta$ a flat (improper) prior, and use $d_1 \\exp(\\alpha + \\beta Z_i), \\, d_1$ as parameters for the beta distribution that governs $b1_i$, and $d_2 \\exp(\\gamma + \\delta Z_i), \\, d_2$ for the beta distribution that governs $b2_i$. Note that with this choice of parameters for the beta distributions, the log odds of the expectation $b1_i$ and of $b2_i$ each depend linearly on $Z_i$ - that is, \n",
"$$\\log \\dfrac{\\mathbb{E}(b1_i)}{1- \\mathbb{E}(b1_i)} = \\alpha + \\beta Z_i.$$\n",
"\n",
"### The model with covariates\n",
"\n",
" $p(\\alpha), p(\\beta), p(\\gamma), p(\\delta) \\propto 1$ \n",
" \n",
" $d_{1} \\sim Exponential(\\lambda)$\n",
" \n",
" $d_{2} \\sim Exponential(\\lambda)$\n",
" \n",
" $b1_i \\, \\mid \\, d_1, Z_i, \\alpha, \\beta \\sim Beta(d_1 \\exp(\\alpha + \\beta Z_i), \\, d_1)$\n",
" \n",
" $b2_i \\, \\mid \\, d_2, Z_i, \\gamma, \\delta \\sim Beta(d_2 \\exp(\\gamma + \\delta Z_i), \\, d_2)$ \n",
" \n",
" $T'_i \\, \\mid \\, b1_i, b2_i, X_i \\sim Binomial(N_i, \\theta_i)$"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<img scr=\"EI_graphical_model_cov.png\"> "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<img src=\"EI_graphical_model_cov.png\"> "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Examples with code"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"There is a nice R package for all this and more [1]. Let's see how we can make it happen in Python!"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import seaborn as sns\n",
"import matplotlib.pyplot as plt\n",
"import pymc3 as pm\n",
"import pandas"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def ei_two_by_two(X, T, N, lmbda):\n",
" Tprime_obs = T * N\n",
" p = len(X) #number of populations\n",
" with pm.Model() as model: \n",
" c_1 = pm.Exponential('c_1', lmbda)\n",
" d_1 = pm.Exponential('d_1', lmbda)\n",
" c_2 = pm.Exponential('c_2', lmbda)\n",
" d_2 = pm.Exponential('d_2', lmbda)\n",
" \n",
" b_1 = pm.Beta('b_1', alpha=c_1, beta=d_1, shape=p)\n",
" b_2 = pm.Beta('b_2', alpha=c_2, beta=d_2, shape=p)\n",
" \n",
" theta = X * b_1 + (1 - X) * b_2\n",
" Tprime = pm.Binomial('Tprime', n=N , p=theta, observed=Tprime_obs) \n",
" return model"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1. A toy example\n",
"Let's imagine we have as four voting precincts, each with 100 people in them. The percentage of Group 1 present in each precinct is 10%, 20%, 30% or 40% respectively ($X$). The percentage of each precinct voting for some political candidate A is 11%, 18%, 34% or 40% respectively ($T$).\n",
"\n",
"It looks like X and T are tracking pretty closely. Does this mean that members of Group 1 overwhelmingly support candidate A, while members of Group 2 do not? (What other explanations can you imagine?)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"X = np.array([.1, .2, .3, .4])\n",
"T = np.array([.11, .18, .34, .4])\n",
"N = np.array([100, 100, 100, 100])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we set up our model and obtain MCMC samples from the posterior, using PyMC3:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Auto-assigning NUTS sampler...\n",
"Initializing NUTS using jitter+adapt_diag...\n",
"100%|██████████| 1000/1000 [00:12<00:00, 25.20it/s]/Users/colin/projects/pymc3/pymc3/step_methods/hmc/nuts.py:451: UserWarning: The acceptance probability in chain 0 does not match the target. It is 0.679785112819, but should be close to 0.8. Try to increase the number of tuning steps.\n",
" % (self._chain_id, mean_accept, target_accept))\n",
"/Users/colin/projects/pymc3/pymc3/step_methods/hmc/nuts.py:467: UserWarning: Chain 0 contains 46 diverging samples after tuning. If increasing `target_accept` does not help try to reparameterize.\n",
" % (self._chain_id, n_diverging))\n",
"\n"
]
}
],
"source": [
"lmbda = 0.5 #chosen to match [King, 1999]\n",
"\n",
"toy_model = ei_two_by_two(X, T, N, lmbda)\n",
"\n",
"with toy_model:\n",
" toy_trace = pm.sample()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can use PyMC3 to plot the traces:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAANYCAYAAADZn0yoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nk99QrpCg0AAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10b3e1e10>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.traceplot(toy_trace);"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Having obtained our samples, we access some statistics of interest. (Note how broad the 95% HPD interval is for the $b1_i$ !) "
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"b_1 mean across districts: 0.8673846099\n",
"\n",
"b_1:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.866 0.161 0.013 [0.495, 1.000]\n",
" 0.850 0.148 0.012 [0.545, 1.000]\n",
" 0.868 0.187 0.016 [0.436, 1.000]\n",
" 0.886 0.139 0.012 [0.572, 1.000]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.458 0.792 0.938 0.992 1.000\n",
" 0.507 0.767 0.887 0.980 1.000\n",
" 0.297 0.810 0.956 0.996 1.000\n",
" 0.534 0.848 0.943 0.985 1.000\n",
"\n",
"\n",
"b_2:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.019 0.031 0.002 [0.000, 0.094]\n",
" 0.024 0.042 0.003 [0.000, 0.111]\n",
" 0.078 0.112 0.010 [0.000, 0.333]\n",
" 0.057 0.094 0.008 [0.000, 0.268]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.000 0.000 0.001 0.025 0.109\n",
" 0.000 0.000 0.001 0.031 0.145\n",
" 0.000 0.000 0.016 0.131 0.420\n",
" 0.000 0.000 0.001 0.082 0.321\n",
"\n"
]
}
],
"source": [
"print(\"b_1 mean across districts: \"+ str(toy_trace.get_values('b_1').mean()))\n",
"\n",
"pm.stats.summary(toy_trace, varnames=['b_1', 'b_2'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 2. An example with data simulated from the generative model\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Simulate some data from the model"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"X = np.array([.1, .2, .3, .4, .5, .6, .7, .8, .9])\n",
"N = np.array([100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 100.0])\n",
"p = len(X) # Number of districts\n",
"\n",
"c_1= 12.0\n",
"d_1 = 3.0\n",
"c_2= 6.0\n",
"d_2 = 8.0\n",
"\n",
"b_1_sim = []\n",
"b_2_sim = []\n",
"theta_sim = []\n",
"Tprime_sim = []\n",
"\n",
"for i in range(p):\n",
" b_1_sim.append(np.random.beta(c_1, d_1))\n",
" b_2_sim.append(np.random.beta(c_2, d_2))\n",
" theta_sim.append( X[i] * b_1_sim[i] + (1 - X[i]) * b_2_sim[i])\n",
" Tprime_sim.append(np.random.binomial(N[i], theta_sim[i]))\n",
"\n",
"T = np.array(Tprime_sim / N)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Obtain MCMC samples from the posterior"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Auto-assigning NUTS sampler...\n",
"Initializing NUTS using jitter+adapt_diag...\n",
" 99%|█████████▉| 988/1000 [00:05<00:00, 216.24it/s]/Users/colin/projects/pymc3/pymc3/step_methods/hmc/nuts.py:467: UserWarning: Chain 0 contains 21 diverging samples after tuning. If increasing `target_accept` does not help try to reparameterize.\n",
" % (self._chain_id, n_diverging))\n",
"100%|██████████| 1000/1000 [00:05<00:00, 191.28it/s]\n"
]
}
],
"source": [
"lmbda = 0.5 #chosen to match [King, 1999]\n",
"\n",
"sim_model = ei_two_by_two(X, T, N, lmbda)\n",
"\n",
"with sim_model:\n",
" sim_trace = pm.sample()"
]
},
{
"cell_type": "code",
"execution_count": 67,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#TO DO:\n",
"table of results"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 3. An example with real voting data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we apply this method of ecological inference to data in [King 1997, 1990], as distributed in [1]. Our goal is to reproduce this example using PyMC3. While we use the data and model of [King 1999],[6], the computational method of sampling differs; we sample with NUTS in PyMC3[5] while [King 1999] used a Metropolis method to obtain the MCMC samples.\n",
"\n",
"The data represents $p$=268 precincts in 1968. We know for each district the percentage the voting age popoulation that is black ($X_i$) and white, the percentage of voting age members of the district who are registered to vote ($T_i$), and size of the total voting age population ($N_i$). \n",
"\n",
"This data set is useful because it also contains 'ground truth' measurements of the proportion of black voting age members of the population who are registered to vote and the proportion of white voting age members of the population who are registered to vote. We can compare the results of the inference to these measurements."
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"#Load the data\n",
"data = pandas.read_csv('ei_example.csv')\n",
"X = np.array(data.x)\n",
"N = np.array(data.n)\n",
"T = np.array(data.t)\n",
"Tprime = np.array(T * N)\n",
"p = len(X)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We set up and sample from the model:"
]
},
{
"cell_type": "code",
"execution_count": 62,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Auto-assigning NUTS sampler...\n",
"Initializing NUTS using ADVI...\n",
"Average Loss = 3,163.8: 15%|█▌ | 30280/200000 [00:18<01:40, 1685.60it/s] \n",
"Convergence archived at 30300\n",
"Interrupted at 30,300 [15%]: Average Loss = 1.8151e+05\n",
"100%|█████████▉| 999/1000 [07:53<00:00, 1.99it/s]/Users/karin/anaconda/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:448: UserWarning: Chain 0 reached the maximum tree depth. Increase max_treedepth, increase target_accept or reparameterize.\n",
" 'reparameterize.' % self._chain_id)\n",
"/Users/karin/anaconda/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:456: UserWarning: Chain 0 contains 59 diverging samples after tuning. If increasing `target_accept` does not help try to reparameterize.\n",
" % (self._chain_id, n_diverging))\n",
"100%|██████████| 1000/1000 [07:53<00:00, 2.11it/s]\n"
]
}
],
"source": [
"lmbda = 0.5 #chosen to match [King, 1999]\n",
"\n",
"voter_model = ei_two_by_two(X, T, N, lmbda)\n",
"\n",
"with voter_model:\n",
" voter_trace = pm.sample()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let us look at posterior samples of $\\dfrac{c_1}{c_1 + d_1}$, the mean of the beta distribution which governs the $b1_i$. We compare them to samples of $\\dfrac{c_2}{c_2 + d_2}$."
]
},
{
"cell_type": "code",
"execution_count": 66,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x11e887790>"
]
},
"execution_count": 66,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD3CAYAAAAALt/WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x127839c10>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.distplot(voter_trace.get_values('c_1')/(voter_trace.get_values('c_1') + voter_trace.get_values('d_1')));\n",
"sns.distplot(voter_trace.get_values('c_2')/(voter_trace.get_values('c_2') + voter_trace.get_values('d_2')));"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"A look at these graphs certainly suggests a marked difference in the rates of voter registration between black (samples in blue) and white (samples in green) voters in these 268 sampled counties in 1968.\n",
"\n",
"We can also look at the posterior means of the the $b1_i$ and $b2_i$ for each county, and compare our inferred values."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"TODO: table of results - posterior means of the the $b1_i$ and $b2_i$ vs. measured 'truth'"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 4. An example with real voting data and covariates"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We modify the previous 1968 voting rate example by providing a covariate $Z$. For the sake of an easy example with the same data set, and to parallel [King 1999], we use $Z = X$ as our covariate. This allows us to consider the question of whether there is some effect of $Z_i = X_i$, the fraction of voters in a precinct who are black, on the rates of black and white voter registration. Using different covariate(s) (for example, spatial variables) would also be interesting."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def ei_two_by_two_cov(Z, X, T, N, lmbda):\n",
" with pm.Model() as covariate_model: \n",
" p = len(X)\n",
" Tprime_obs = T * N\n",
" \n",
" alp = pm.Flat('alp')\n",
" bet = pm.Flat('bet')\n",
" gam = pm.Flat('gam')\n",
" dlt = pm.Flat('dlt')\n",
" \n",
" d_1 = pm.Exponential('d_1', lmbda)\n",
" d_2 = pm.Exponential('d_2', lmbda)\n",
"\n",
" c_1 = d_1*pm.math.exp(alp + bet*Z)\n",
" c_2 = d_2*pm.math.exp(gam + dlt*Z)\n",
" \n",
" b_1 = pm.Beta('b_1', alpha = c_1, beta = d_1, shape=p)\n",
" b_2 = pm.Beta('b_2', alpha = c_2, beta = d_2, shape=p)\n",
" \n",
" theta = X * b_1 + (1 - X) * b_2\n",
" Tprime = pm.Binomial('Tprime', n=N , p=theta, observed=Tprime_obs)\n",
" \n",
" return covariate_model"
]
},
{
"cell_type": "code",
"execution_count": 73,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Auto-assigning NUTS sampler...\n",
"Initializing NUTS using ADVI...\n",
"Average Loss = 3,118.6: 15%|█▍ | 29599/200000 [00:21<02:24, 1178.41it/s] \n",
"Convergence archived at 29600\n",
"Interrupted at 29,600 [14%]: Average Loss = 1.8586e+05\n",
"100%|██████████| 1000/1000 [13:17<00:00, 1.55it/s]/Users/karin/anaconda/lib/python2.7/site-packages/pymc3/step_methods/hmc/nuts.py:456: UserWarning: Chain 0 contains 8 diverging samples after tuning. If increasing `target_accept` does not help try to reparameterize.\n",
" % (self._chain_id, n_diverging))\n",
"\n"
]
}
],
"source": [
"Z = X # As a simple example, use X as the covariate\n",
"\n",
"cov_model = ei_two_by_two_cov(Z, X, T, N, lmbda) \n",
"\n",
"# Generate MCMC samples from the model\n",
"with cov_model:\n",
" cov_trace = pm.sample()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Perhaps we are interested in $\\delta$, which governs the relationship between $Z_i$ and the beta distribution that governs $b2_i$. Recall that under the model, \n",
"$$\\log \\dfrac{\\mathbb{E}(b2_i)}{1- \\mathbb{E}(b2_i)} = \\gamma + \\delta Z_i.$$ \n",
"\n",
"Note that the posterior distribution of $\\delta$ is centered near 0, with a 95% HDP interval of [-1.437, 1.458]. So the log odds of the expected mean of $b2$ does not appear to depend linearly on $Z$."
]
},
{
"cell_type": "code",
"execution_count": 81,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"-0.0130718575936\n",
"\n",
"dlt:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" -0.013 0.730 0.046 [-1.437, 1.458]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" -1.637 -0.501 0.024 0.487 1.357\n",
"\n"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD3CAYAAAAALt/WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xl0W+d55/HvBUAABAHu4L5reanNkmzZlh05lhM7jp06\ndtN00rpNpzl1p+ly5kzcmTbtrD3TmZ5O6qbptJm2M03TLYmzuXGb2HESyZtsObasfXklipS4ivsC\nbljv/AHSpm2JBEkQFxd4Puf4WOQFwOdKFz++eO+7GKZpIoQQwr4cVhcghBBifSTIhRDC5iTIhRDC\n5iTIhRDC5iTIhRDC5lyZ/oHDw6FVD5MpK/MxPj67EeVYLpfPDXL7/OTc7MuO5xcMBowbHbNFi9zl\nclpdwobJ5XOD3D4/OTf7yrXzs0WQCyGEuDEJciGEsDkJciGEsDkJciGEsDkJciGEsDkJciGEsDkJ\nciGEsDkJciGEsDkJciGEsLmMT9EXwmrPn+hb9XMO7qnfgEqESA9pkQshhM1JkAshhM1J14oQG0i6\ncUQmSItcCCFsToJcCCFsToJcCCFsToJcCCFsToJcCCFsToJcCCFsToJcCCFsToJcCCFsToJcCCFs\nToJcCCFsToJcCCFsbsW1VpRSDuCLwG4gDDymte5YcvwLwAEgtPCth7XWkxtQqxDvsJZ1TKxkmibR\neIICpwPDMKwuR+SQVBbNegTwaq3vUErtB54AHl5y/Bbgfq31yEYUKIQdzYVjXO6f5MSlEYbG55iY\nDhOOxjFNcDkNAj43VWWFbG8pI+BzW12usLlUgvwA8CyA1vqoUmrf4oGF1voW4K+UUtXAX2utv7Qh\nlQqRpeKJBMMT81y5NkVH7ySXeifpHZ7GNN9+jL+wgICvAHeBk9n5GFMzEcZDYS72TNBaW8y+9iq8\nbqd1JyFsLZUgLwaWdpXElVIurXUMKAL+N/DHgBM4rJR6Q2t96kYvVlbmw+Va/QUbDAZW/Ry7yOVz\ng407v4Dfm7bXmpwO0zs0zcDIDNNzUebCMWLxBE6HgcNh8OrZQVxOBwUuBy6nA9M0k4E8G2FobJZ4\n4u3UdrscbG+tYFtLOXPzMWoqfHg973yrJRImHb0THLswRGf/FGOhMA8daKO4yJ0110O21LFRcun8\nUgnyKWDpGTsWQhxgFviC1noWQCl1iGRf+g2DfHx8dtVFBoMBhodDKz/QhnL53GBjzy80Pb/u17g2\nOsuJjmT3x1JulwOn00HENIknTEIzEWJxk8SSZrbLaeDzuGipCVBT7qM+6GdLQwnNNQFczuQ4gudP\n9BGNxohGY7xbbXkhH7mjiTcvDnO2a5xvHrrIB29pyIrrQa7L7LPcL55UgvwI8BDw9YU+8tNLjm0F\nnlRK7SU5AuYA8LdrL1WIzJiPxHj51AD9I8mGRV2lj4YqP7XlPgI+Nw7HO29GLm72kEiYxBMJTBPq\naksYGZleVx2GYXCLqsLnKeD1C0McerOPD93ahL+wYF2vK/JLKkH+FHCfUuoVwAA+pZR6HOjQWj+t\nlPp74CgQBf5Oa31248oVYv3GpuY5/GYfM/Mxait87N1aSWVJYUrPdTgMHI5k12A6R55saykjGk9w\n4tIIf/O98/zGx3bJyBaRshWDXGudAD79rm9fWHL8c8Dn0lyXEBtiYHSGw2/2EYub7NlSya628qwJ\nzJ1t5QyMznD80giH3uzjg7c0WF2SsAmZECTyxsR0mOeP95NIwMG9ddy0qSJrQhzAYRjcdVMd/sIC\nnjzUwfDE3MpPEgIJcpEn5sIxDh3rIxpLcOeuapqqs3PEgs/r4tF7txCLJ/inlzqtLkfYhAS5yHmm\nafLiiX6m56Ls3lxBW12J1SUt67bt1TRV+Tl6dpDuQXuNrBDWkCAXOe9izwSD43M0Vvm5aVOF1eWs\nyGEYfPzgJkzgWy9Iq1ysTIJc5LSZ+Shv6hHcLge3b6/Oqj7x5exoLae9qZTTnaPo7nGryxFZLpXh\nh0LYkmmaHD07SDSe4I6d1fi89rjcFxcDa60r5kL3BP/wg4spjWBZHOsu8o+0yEXO6hueoW94hppy\nH5vrs7tf/HqCpYUES730Dc8Qmo1YXY7IYhLkIieZpsmJjuSCnLduq7JNl8q7qaYyAHT3hMWViGwm\nQS5yUvfgNGNTYVprA5QFPFaXs2bNNQG8bicdvZNEYwmryxFZSoJc5JyEaXLi0giGAbs3V1pdzro4\nHQZbG0uJxBJ0DUxZXY7IUhLkIudcGZhicibCpvoSiovsv2nD1sZSDCPZvWIuXeRciAUS5CKnmKbJ\n2a5xDANuasv+MeOp8HldNFb5GQ+FGQ+FrS5HZCEJcpFTBsfmGA+FaaoO4PflzlKwrbXFANK9Iq5L\nglzklPNXk5NntjWXWVxJejUEiyhwOegaCEn3ingPe8yQECIFodkIPUPTVBR7CZambxs4eHuSzlIB\nvzctuxSlwul00FwdoKNvksGxOWoqfBn5ucIepEUucsbiWOttLaW2HTe+nNa65IqNndK9It5Fglzk\nhFg8waXeSQo9Tppriq0uZ0NUl/vweVx0XwsRT8iYcvE2CXKRE7oHp4nGEmyuL8HpyL3WOCRXRWyp\nDRCJJegbnrG6HJFFJMhFTrjcNwnAJhuuqbIazTXJ7pWeofVt+ixyiwS5sL2ZuSgDo7MESwtzYgLQ\ncipLvBR6XPQOzZBIyOgVkSRBLmzvcn/y5t/m+tzsG1/KMAwaq/yEo3GGxmVPT5EkQS5szTRNLvdN\n4nQYNNdm5z6c6dZU7Qeke0W8TYJc2NrwxByh2SjNNQHcLqfV5WREdbmPApeD7kGZHCSSJMiFrXUN\nJDcnbqvL/W6VRU6HQUOwiJn5GGOy9opAglzYWMI0uXothKfASU15fs10bKpeGL0yKN0rQoJc2NjQ\n2BzzkThN1X4cOTp2/EbqKotwGAa9wxLkQoJc2NiVa8lulZY8ucm5VIHLQVV5IWNTYebCMavLERaT\nIBe2FE8k6B4M4XU7qS7Lr26VRfWVRQD0j8gsz3wnQS5s6WL3RN52qyxaDHKZri8kyIUtvX5hCICW\nHF0gKxUlfjc+r4v+0RkSMgwxr624HrlSygF8EdgNhIHHtNYd13nMd4HvaK3/YiMKFWJRwjR58+Iw\nXreTqvJCq8uxjGEY1FcWcal3ktHJzKyLLrJTKi3yRwCv1voO4LPAE9d5zO8DubUli8hanf1TTM1G\naQj6ceTguuOrUR+U7hWRWpAfAJ4F0FofBfYtPaiU+jiQWHyMEBvt+KVhABoXpqrns5pyH4YhNzzz\nXSpbvRUDk0u+jiulXFrrmFJqJ/Ao8HHgv6TyA8vKfLjWMJU6GMzdIWa5fG6Q/vM73TmGu8DJlqZy\nClzW3uYJ+NO7pdxa1FYU0T8yg7vQTYnfk7bXlevSPlIJ8ilg6Rk7tNaLA1d/AagHDgEtQEQpdUVr\nfcPW+fj47KqLDAYDDA+HVv08O8jlc4P0n9+1sVl6h6bZu6WS+fkIVvYMZ3LPzuVUlxfSPzLDi290\ns39HTVpeU67L7LPcL55UmjNHgAcBlFL7gdOLB7TWv6W1vl1rfRD4MvDHy4W4EOu12K2yZ0ulxZVk\nj8VhiKc7xyyuRFgllSB/CphXSr0CfB74jFLqcaXURze2NCHe6/ilEQwDdm+WIF9UFvBQ6HFypmtU\nhiHmqRW7VrTWCeDT7/r2hes87r+lqSYhrmtqNsLl3kk2N5RQ7MvtnYBWwzAM6iqLuNw3RfdgKK/H\n1ucrmRAkbONM5ygmsEda4+/xVvfK5VGLKxFWkCAXtnFqIaRu2lRhcSXZp7aiCMOA013ST56PJMiF\nLcQTCc52jVFR7KFuofUp3uZxO2mrK+Zy3yQz81GryxEZJkEubKGzf4qZ+Ri7NlVi5PlszhvZ1VaB\nacJZaZXnHQlyYQtvdau0SbfKjexa+LuRIM8/EuTCFk5dHsXlNNjWLEv63EhzdYAir4uzV8ZkU+Y8\nI0Eust54KEzP0DSqqQyPe/XLO+QLh8Nge0s5Y1Nhro2tfga1sC8JcpH1TnfKaJVU7WwtB+CMdK/k\nFQlykfUWg3yX9I+vaMdCkEs/eX6RIBdZLZEwOX9lnMoSL9Vl+buJRKrKi73UVvi40D1ONJawuhyR\nIRLkIqt1XZtiNhxje0u5DDtM0Y7WciLRBB19kys/WOQECXKR1c4tdBEs9v2Kle2U7pW8I0EustrZ\nrjEMoF2GHaZMNZbhdBgS5HlEglxkrblwjMv9U7TUFuMvLLC6HNvwuJ1saSjh6mCIqdmI1eWIDJAg\nF1lLd08QT5jsaJXW+Gotjl45d0Va5flAglxkrbMLIbSjRfrHV2tnq0zXzycS5CJrnbsyhqfAyab6\nEqtLsZ3Gaj8BXwFnu2S6fj6QIBdZaWxqnoHRWdqbSnE55TJdLYdhsKOlnInpCP0jM1aXIzaYvENE\nVlrsEtguww7XTGZ55o8V9+wUIhOeP9H3jq9fPNEPwOx89D3HRGq2L9xbOHNljA/d1mRxNWIjSYtc\nZB3TNBkYncXndVFcJJssr1VZwEN9sIiL3RNEY3GryxEbSIJcZJ2xqTDhaJy6iiKZlr9OO1rKicQS\nXOyR6fq5TIJcZJ3+0eTNudpKn8WV2N+uhaV/F3dYErlJglxknYGR5KYItRUS5Ou1taEUT4GTU5dH\nrC5FbCAJcpFVorEEQ+NzlBd78LrlXvx6FbgcbG8pY3B8jkHZNShnyTtFZJWh8VkSpkltRZHVpdjO\njUb3eBe2x/v2i51sa3nncgcH99RveF1i40mLXGSV/oVulTrpH0+b+mDyl2Lv8LTFlYiNIkEussrA\n6AxOh0GV7AaUNj5vAWUBD4Njc7JrUI6SIBdZY3Y+xsR0hOpyH06HXJrp1BAsImGaDIzKdP1cJO8W\nkTUWQ6ZORqukXX3QD0DfsAR5LlrxZqdSygF8EdgNhIHHtNYdS47/OvCLgAn8kdb66xtTqsh1i4s7\n1VbKjc50qyz14ilw0jcyg2maMtEqx6TSIn8E8Gqt7wA+CzyxeEApVQn8KnAn8EHgCaWUXCFi1Ran\n5Rd6nJT6ZVp+ujkMg7pK30L3VdjqckSapTL88ADwLIDW+qhSat/iAa31iFJqj9Y6ppRqAea11ssu\nflxW5sPlcq660GAwsOrn2EUunxukdn6ROMxH4qjmMooD9rnRGfB7rS4hZZsby+gaCDE8GaapthRY\n/t9Grkv7SCXIi4GlCzXElVIurXUMYCHEfwP4PeBPV3qx8fHVT0oIBgMMD4dW/Tw7yOVzg9TPr6Nn\nPPn4Eg+h6fmNListAn6vbWoFKPe7MYDLvZNsbUhu1nGjfxu5LrPPcr94UulamQKWvoJjMcQXaa3/\nDKgF3q+UumctRYr89lb/uEwE2jAet5PK0kJGJuYIR2Q1xFySSpAfAR4EUErtB04vHlBJ317oF4+S\nvBkqA1XFqkSicYbG5ygLeCj0yGTjjdQQLMIE2TUox6QS5E8B80qpV4DPA59RSj2ulPqo1loDJ4FX\ngVeAo1rrFzauXJGLLvVOEk+YskhWBizO8uyTIM8pKzZ/tNYJ4NPv+vaFJcd/j2T/uBBrcvZKciuy\nOhl2uOHKAh58Hhd9wzMkZFPmnCETgoTlznaN4ZBp+RlhGAYNVUWEo3GGx+esLkekiQS5sNTkTISe\noWmqywpxOeVyzITGquTYhe5BWUQrV8g7R1jq3EK3ivSPZ05NhY8Cl4OeoWlM6V7JCRLkwlLnuqR/\nPNOcDoP6YBHTc1F6hqRVngskyIVlTNPkzJUxAr7kMqsic5qqkotovXlx2OJKRDpIkAvL9I3MMDkd\nYXtLuSzilGH1QT8OhyFBniMkyIVlznQmu1V2tZVbXEn+KXA5qK3w0Ts8w9Aals0Q2UWCXFjmdOco\nADtaKyyuJD81VSe7V45Jq9z2JMiFJebCMS72TNBcE6CkSJattUJjlR+HYfDGhSGrSxHrJEEuLHGh\ne5x4wmRXm7TGreJ1u9jWsrC07YRMDrIzCXJhidPSP54Vbm2vAuB1aZXbmgS5yDjTNDl9eRSfx0Vb\nXbHV5eS1m7cGcToMXj8vQW5nEuQi466NzTI6Nc+O1nKcDrkEreQvLGB7SzlXB0MMyugV25J3kci4\n05eTo1V2SrdKVljsXpGbnvYlQS4ybnHYodzozA57t1bidBi8dk6C3K4kyEVGhSNxdM8ETVV+Sv0y\nLT8bFHkL2NVWQe/wNH3DsvaKHUmQi4y60D1OLG6yU1rjWWX/jmoAjp4btLgSsRYS5CKj3u5Wkf7x\nbLJ7cyVet5OjZwdl5yAbkp1uRUad6Ryj0ONkU32J1aUI4PkTfW/9ub6yiMv9U3zjcAebm8oJTc/f\n8HkH99RnojyRImmRi4wZHJtlaGKO7c3lshtQFmpdGNPfNTBlcSViteTdJDLm1GK3yibpH89GNRU+\nCj1OrlwLEU8krC5HrIIEuciYxWVrd7ZK/3g2chgGrbXFRKIJrkir3FYkyEVGhCNxLnSPUx8sorzY\na3U54gYW712cX9iCT9iDBLnIiLNXxojGEuzZXGl1KWIZZQEPFSVeuq+FmJ2PWl2OSJEEuciI45eS\nmxfs3RK0uBKxki31JZjA5T7pXrELCXKx4RIJk5Mdo5T43bTUBqwuR6ygpTaAy2nQ0TeJKWPKbUGC\nXGy4jr5Jpuei7N1ciUM2Wc567gInmxpKCc1GGRyTDSfsQIJcbLjFbpU90q1iG9tbkiOLOvomLa5E\npEKCXGwo0zQ5fmkEj9vJtuYyq8sRKaqtLCLgK+DqtRCRaNzqcsQKJMjFhuoZDDE0Pseu1nIKXHK5\n2YVhGGxuKCGeMGWmpw2suNaKUsoBfBHYDYSBx7TWHUuOfwb4mYUvv6e1/r2NKFTY06unBwAZrWJH\nm+pKOHFphI7eSVSTfJrKZqk0kR4BvFrrO4DPAk8sHlBKtQE/B9wJ7Ac+pJS6aSMKFfb08sl+XE6D\n3TJ+3HZ8Xhf1lUWMToUZm7rxAlrCeqmsfngAeBZAa31UKbVvybEe4MNa6ziAUqoAWPZfvKzMh8vl\nXHWhwWDuDlvL1XPrGQxxZWCK23fU0Ny4fIsu4LfnbE+71p2KgN/Lri1BeodnuDo0Q3Nd6VvHcuGa\nzYVzWJRKkBcDS29dx5VSLq11TGsdBUaUUgbwOeC41vrici82voYNXoPBAMPDoVU/zw5y+dyee7UL\ngJtay1c8x+WWTM1WAb/XlnWnYvHcKvxuvG4n+uoYu1rL3lq10u7XrB3fd8v94kmla2UKWPoKDq11\nbPELpZQX+MeFx/zaGmsUOeiNC0MUuBzs2SLdKnblcBhsri8hEk1w9Zq9gi+fpBLkR4AHAZRS+4HT\niwcWWuLfAU5qrX9lsYtFiP6RGXqHZ7hZVVHokf1L7GxrUykGcOHquMz0zFKpvMOeAu5TSr0CGMCn\nlFKPAx2AE7gb8CilHlh4/O9orV/dkGqFbbxxIbkj+4HddRZXItbLX1hAQ5WfnqFpRibnCZYWWl2S\neJcVg1xrnQA+/a5vX1jy59y92yNW7fkTfZimyeHjfTgcBlMzkXdsJybsqb25lJ6haXT3hAR5FpIZ\nGiLtRqfmmZyJ0Fjlx12w+hFKIvvUlPsoKXJzZSDEXDi28hNERkmQi7RbXP5008IekML+DMNANZWS\nME06emX9lWwjQS7SKp4wuTIQwut2UldZZHU5Io3a6otxOQ10z4Ts6ZllJMhFWvUNTxOOxmmtLcbh\nkCVrc4nb5WRTfQmz8zFOXBqxuhyxhAS5SKvO/oVulXrpVslFqik5u/NHx3otrkQsJUEu0iY0G6F3\naJpSv5uygMfqcsQGKPV7qCn3caF7gr7haavLEQskyEXavHx6gIQJmxtKMGQnoJzV3pxslR96U4aV\nZgsJcpEWCdPk+eN9OB0Gm+pKrC5HbKCGoJ+KYg9HzgwwPRe1uhyBBLlIk7NdYwxPzNNSG8DjlrHj\nuczhMLh3XyORaILDx6VVng0kyEVaHF74mL14M0zktvfvrqPQ4+RHx3qJxmSJJatJkIt1G52c5+Tl\nEVpqAlSWyPTtfFDocXFwTz1TMxFePTtodTl5T4JcrNvh432YJtyzt97qUkQG3buvEafD4NnXuknI\nqoiWkiAX6zI7H+Pw8V6KfQXcvr3a6nJEBpUFPOzfXs21sVlOdYxaXU5ekyAX6/LCiT7mwnHuu7VR\nFsjKQ/ff3gTAs69dtbiS/CZBLtYsGovz3Os9eN1O6VbJUw1BPzvbyrnYO8nlfllMyyoS5GLNXjlz\njcmZCPfsrcfnLbC6HGGRB25Ltsq//1q3xZXkLwlysSaxeIJnjnbjchrcd2uj1eUIC7U3l9FU7efY\nxWGG1rC5ulg/CXKxJi+dGmBoYo73766j1C/rquQzwzD48O1NmCY8++Meq8vJSxLkYtXmIzG+83IX\nngInD72v1epyRBa4tb2KqtJCXj7Vz9jUvNXl5B0JcrFqP3i9h6mZCPff1khJkdvqckQWcDoc/MSd\nLcTiJt89KiNYMk2CXKzK1GyEZ17rxl9YwP0LN7mEALhjZzVVpYW8dFJa5ZkmQS5W5ZvPX2Y+Eueh\n97VQ6HFZXY7IItIqt44EuUjZxZ4JXj41QGOVnw/cLOPGxXtJq9waEuQiJbF4gr999gIG8AsfVjgd\ncumI95JWuTXk3ShS8sxr3QyMznJwb71sHCGWJa3yzJMgFyvqGpji6Ze7KPG7+am726wuR2Q5aZVn\nngS5WNZ8JMZfPX2WeMLksY9sl6n4IiXSKs8sCXKxrK/+8BKD43N8+LYmdrSWW12OsImlrfKnj3RZ\nXU7OkyAXN/TiyX5eOjVAU7Wfj0mXililO3ZWU1dZxEunBugbmbG6nJy2YpArpRxKqb9QSr2qlHpe\nKbX5Oo8JKqUuKqW8G1OmyLSLPRP8/fc1RV4Xv/bITlxO+Z0vVsfpcPDxuzdhmvCt5y9bXU5OS+Xd\n+Qjg1VrfAXwWeGLpQaXU/cBzQE36yxNWGJmc48+fOo1pwq89spOqMp/VJQmb2r25gq2NpZzoGOFi\nz4TV5eSsVIL8APAsgNb6KLDvXccTwL3AWHpLE1aYmonwxJMnCc1GefS+LWxrkX5xsXaGYfDT92wC\n4MlDHbK35wZJZY51MbB064+4UsqltY4BaK1/AKCUSukHlpX5cLlWvyVYMBhY9XPsIlvObXouyu//\n3TEGx2b5qXs284n7t636NQL+9/auXe97uSJfz20112wwGOCuPdd46UQfZ7sn+MC+7FijJ1ved+mQ\nSpBPAUvP2LEY4msxvoaF54PBAMPDobX+yKyWLec2Ox/jT75xks7+SQ7uqePB2xrXVFdo+p1DzQJ+\n73u+lyvy+dy+8YMLq3q9YImHApeDLz19ls01AcvX6cmW991qLPeLJ5WulSPAgwBKqf3A6fSUJbJF\naDbC5752nI6+SfZvr+bnP6QwDMPqskQO8RcW8MDtTUzORPjuqzJJKN1SCfKngHml1CvA54HPKKUe\nV0p9dGNLE5kwHgrzv75ynKvXQtx1Uy2P/cR2HA4JcZF+D+xvprzYw3OvdzM4JlvCpdOKn2+01gng\n0+/69ns+V2mtW9JUk0iz50/0Xff7o1PzHDrWx1w4RntzKS21AV481f/W8YN7ZIVDkT6eAief+MAW\n/s8/neHvn9P85if2yCe/NJHBwXmqezDE91/rZi4c4xYV5Nb2KnlTiQ23TwXZ2VbOuSvjvHZ+0Opy\ncoYEeZ5JJEyO6SGeP55seR/cW8eO1nIJcZERhmHw8x9SFLgcfO1HHczOR60uKSfIFi95ZHouypFT\nAwyOzxHwFXBwbz1lAc8NH3+jLhkh1qOqtJCPvq+Fb73QydcPX+YXH2i3uiTbkxZ5HjBNk87+Sf75\nyBUGx+dorvbzkTublw1xITbS/bc10RAs4sWT/Zy9InMJ10uCPMeNTc1z+Hg/L5+6hmma3LGzhvfv\nqcO9hklZQqSLy+nglz6yHYdh8OXvnWcuvOapKQLpWslZsXiCQ2/28dRLnYQjcarLC7lzZw0Bn9vq\n0kSeul5X3Y62ck5fHuUL3zzJ/h3XX65JRk+tTII8x5imycmOUZ483MHg2CxFXhe37KxhU32x3NAU\nWeemTeX0DIa42DNJQ9BPQ5Xf6pJsSYI8h3QPhnjyUAfnr47jMAzuubmehw+08ubFYatLE+K6nA4H\nd+2u5buvdnPk9DUeel8LPq/E0mrJ35iN3GgUyex8jBMdI3T0Jtc2qw8WcYsKUur3SIiLrFcW8HLL\n1iCvXxjiyOkB7t3XIJ8eV0mC3MZi8QTnroxzpnOUWNyk1O9mX3sVdZVFVpcmxKq0N5fSPzJD38gM\nZ7vG2NlWYXVJtiJBbkOmaXLlWog39TAz8zG8bif72ivZXF8i66QIWzIMgzt31fAvr1zl+MURKkq8\n1FZIgyRVMvzQZoYn5nj2tW5eOjnAXDjOjtZyHnl/K1sbSyXEha0VelzcvacODHjp5AAzMuszZdIi\nt4nQbIQjpwa43D8FQHO1n5tVUIYTipxSVVbIre1V/Pj8EC8c7+dDtzVaXZItSJBnOdM0OXpukK/+\n8BLTc1HKiz3c2l5Fdbnsoylyk2oqZXRynsv9Uxw5NcAHbm7AITc/lyVBnsVGJuf4u+9rznSO4S5w\nsE8FaW8uky4UkdMMw2D/zmqm56JcHZzm2y908vGDm6wuK6tJkGehRMLkR8d6+faLnYSjyX7wX7hf\nyZoUIm84HQ7u3lvPM0ev8r2jVyn1u7l3n3Sz3IgEeZbpGZrmy89coGtgiiKvi0/ev407dtTIuFqR\nd7xuJ/fua+DQsT6+8sNLeN0uDtxUa3VZWUmC3CKLk3sWN7mNxxOcujzKma4xTBNaawPsa68iEkvw\nwsn+FV5NiNwU8Ln5zZ/Zwx/+45v8zTPncRc4uG1btdVlZR0ZfpgFro3O8s9HrnC6cwyfx8UHb6nn\nrt11lu80LkQ2aAj6efwTe/AUOPnLp8/yojRs3kOC3EKz8zGee+0qz73ew9RslG3NZXz0QCv1QVk4\nSIilWmuL+Q8/u5cibwFffuYC3/9xt9UlZRVp8lkgnkhw/so4JzpGiMYSVJR4uX17NZUlXqtLEyJr\ntdYW89sP7yjbAAAHS0lEQVQ/dzN/9LXjPHmog6HxOX723i24nNIelSDPINM0OXdlnK8f7qBnaBp3\ngYO7b26gMeiTcbJCpKC+soj/+Mlb+NNvnubw8T6ujc3y6Yd35P3EOAnyDOnom+SpFzs5f3UcgE31\nxdyiggTL/YSm5y2uTgj7qCwp5Hc/eTP/95/PcfzSCP/1Sz/mlx/awbbmMqtLs4wE+QZKmCZnu8Z4\n5uhVLnRPALCrrYKfuruNzoEpi6sTwr68bhe//rFdPHP0Kk+92MUfffU499/exMMHWvEU5N82hhLk\n63S9NcLnwjE6eie51DvJ9Fxy4Z+6yiJ2tZVTXe6TEBciDRyGwUfuaKG9qYy/fPosz77WzRsXhvjk\n/YpdebYMrgR5msQTCQZGZunom6RnaBrTBJfTYHN9CaqplAq5kSnEhthUX8J//6Xb+c6RLp77cQ+f\n//pJdraV8/G7N9FUHbC6vIyQIF+HSDRO92CIq9dC9A7PEI0lACgLeNjaWEJrbTHuPPyYJ0SmedxO\n/tU9m9m/vZonD3VwpnOMM51j3LI1yIf3N7GprsTqEjeUBPkqjU3Nc7pzlNOdY5ztGiMcjQNQ5HWx\npaGEltoAFcVemVIvRJrcaIvDG9nXHqQ+WMTlvkmOXRzm2MVhNteXcPeeOva1V+VkH7oE+QrmwjEu\n9U5y4eo4pztH6RuZeetYVVkhVaWFNNUEqCj2SHgLkQUMw6Cusoif/eAWdPcEz7zWzenOUTr6JvnK\nDy+yd0uQD97WTEO5lwJXboS6BPkSpmkyNhXm6mCIiz0T6J4JugdDmGbyeIHLwa62Cna1lbNrUwXV\nZb5VtxaEEJlhGAbtzWW0N5cxPDHHy6cGeOXMAK+cucYrZ65R4HKwub6E9uYytjWV0VIbsO3kohWD\nXCnlAL4I7AbCwGNa644lx38Z+BUgBvy+1vpfNqjWtDFNk8mZCCMT8wxNzNI7PEP3YIjuwem3RplA\n8mbllvoStjaVohrL2NJQIn3eQthQsLSQn3x/G4/c1UrnwBTnuid549w1zl8d5/zVcZ4CPAVOGqv9\nNFQWUR/0U19ZRF1lEQFfQdZ/2k6lRf4I4NVa36GU2g88ATwMoJSqAf4tsA/wAi8rpX6gtQ5vRLHj\noTDReAIzYZIwTRIJk3jCxDQhnjCJxuKEowki0TiRWJz5SJzp2SihuSjTc1GmZyOMT0cYmZgjsnBj\ncqmq0kLam0ppqg6wub6Etjq5WSlELjEMg011Jezf3cBD+5sIzUbQ3ROc7x7nYvcEnX1TdPROvuM5\nBS4HpX43ZX4PpQEPpX4PhR7Xwn9OCt0uPG4nLoeB0+nA6TRwORb+73TgdBgYC69T4vdsyHmlEuQH\ngGcBtNZHlVL7lhy7DTiyENxhpVQHcBPweroL/cHrPXz1R5fW/TqFHhe1FUUES70ESwupLC2krsJH\nY1UAn1d6moTIJwGfm33tVexrrwIgGktwbWyWvuFp+kZm6B+ZYSwUZiIU5lLvJOY6f96nHmjnrt11\n6y/8XVJJrmJg6a+ouFLKpbWOXedYCFh2nE8wGFjTZ5RHH9zOow9uX8tTN9RP39dudQlCiDUIBq8/\nxryutoSbM1zLeqXSsz8FLD1jx0KIX+9YAJhIU21CCCFSkEqQHwEeBFjoIz+95NiPgbuUUl6lVAmw\nDTiT9iqFEELckGGay/f6LBm1chNgAJ8iGewdWuunF0at/BuSvxT+p9b6WxtbshBCiKVWDHIhhBDZ\nzZ6j34UQQrxFglwIIWxOglwIIWzOFjNglFJFwFeAMiAC/GutdU4scrIw2ucfSI7JdwOPa61ftbaq\n9FJK/STw01rrR62uZb1WWrIiVyilbgf+UGt90Opa0kUpVQB8CWgBPCSXFHna0qLSxC4t8l8Gjmmt\n308y9H7L4nrS6XHgR1rru4FfBP7c2nLSSyn1BeAPsM+1tpK3lqwAPktyyYqcopT6LeD/kVx2I5f8\nPDCqtb4L+DDwZxbXkza2eHNprf8E+B8LXzaRW5OOPg/85cKfXUCu7cT8CvCrVheRRu9YsoLkOkO5\n5jLwMauL2ADfAP7zwp8Nkgv95YSs61pRSv0S8Jl3fftTWuvXlVKHgF3AfZmvbP1WOLcakp82/l3m\nK1u/Zc7tSaXUQQtK2ijLLVmRE7TW31JKtVhdR7ppracBlFIB4JvAf7K2ovTJuiDXWv818Nc3OPYB\npVQ78F1gU0YLS4MbnZtSahfwNeDfa61fyHhhabDcv1uOWW7JCpHllFKNwFPAF7XWX7G6nnSxRdeK\nUup3lFKfXPhyGohbWU86KaW2k/zI96jW+hmr6xErWm7JCpHFlFLVwHPAb2utv2R1PemUdS3yG/gS\n8LcLH9+dJJcJyBV/QPKm0heUUgCTWuuHrS1JLOMp4D6l1Cu8vWSFsIffJTny7T8rpRb7yh/QWs9Z\nWFNayBR9IYSwOVt0rQghhLgxCXIhhLA5CXIhhLA5CXIhhLA5CXIhhLA5CXIhhLA5CXIhhLC5/w//\nP/0ex3/XUwAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x125358bd0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"print(cov_trace.get_values('dlt').mean())\n",
"\n",
"sns.distplot(cov_trace.get_values('dlt'))\n",
"\n",
"pm.stats.summary(cov_trace, varnames=['dlt'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# References\n",
"\n",
"[1] King, Gary and Roberts, Molly.(2016). ei: Ecological Inference. R package\n",
" version 1.3-3.\n",
" \n",
"[2] King, Gary, Ori Rosen, and Martin A. Tanner. \"Binomial-beta hierarchical models for ecological inference.\" Sociological Methods & Research 28.1 (1999): 61-90.\n",
"\n",
"[3] King, Gary (1997). A Solution to the Ecological Inference Problem. Princeton, NJ: Princeton University Press.\n",
"\n",
"[4] Rosen, Ori, et al. \"Bayesian and frequentist inference for ecological inference: The R× C case.\" Statistica Neerlandica 55.2 (2001): 134-156.\n",
"\n",
"[5] Salvatier J, Wiecki TV, Fonnesbeck C. (2016) Probabilistic programming in Python using PyMC3. PeerJ Computer Science 2:e55 https://doi.org/10.7717/peerj-cs.55\n",
"\n",
"[6] \"Thornburg v. Gingles\". https://ballotpedia.org/Thornburg_v._Gingles\n",
"\n",
"[7] Metric Geometry and Gerrymandering Group https://sites.tufts.edu/gerrymandr/"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "pymc3",
"language": "python",
"name": "pymc3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.1"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
@inikishev
Copy link

congratulations for you have managed to become the owner of the most starred gist that contains the word "uuuh"!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment