Bayesian regression in RJAGS

Bayesian Modeling with RJAGS

Alicia Johnson

Associate Professor, Macalester College

Bayesian regression model

$Y_i$ = weight of adult $i$ (kg)
$X_i$ = height of adult $i$ (cm)

$\;$

Model

$Y_i \sim N(m_i, s^2)$
$m_i = a + b X_i$

$a \sim N(0, 200^2)$
$b \sim N(1, 0.5^2)$
$s \sim \text{Unif}(0, 20)$

Prior insight

Insight from the observed weight & height data

$Y_i \sim N(m_i, s^2)$
$m_i = a + b X_i$

wt_mod <- lm(wgt ~ hgt, bdims)
coef(wt_mod)

(Intercept)         hgt 
-105.011254    1.017617

summary(wt_mod)$sigma

9.30804

DEFINE the regression model

weight_model <- "model{
    # Likelihood model for Y[i]





    # Prior models for a, b, s




}"

DEFINE the regression model

weight_model <- "model{
    # Likelihood model for Y[i]
    for(i in 1:length(Y)) {


    }

    # Prior models for a, b, s




}"

$Y_i \sim N(m_i, s^2)$ for $i$ from 1 to 507

DEFINE the regression model

weight_model <- "model{
    # Likelihood model for Y[i]
    for(i in 1:length(Y)) {
      Y[i] ~ dnorm(m[i], s^(-2))

    }

    # Prior models for a, b, s




}"

$Y_i \sim N(m_i, s^2)$ for $i$ from 1 to 507

DEFINE the regression model

weight_model <- "model{
    # Likelihood model for Y[i]
    for(i in 1:length(Y)) {
      Y[i] ~ dnorm(m[i], s^(-2))
      m[i] <- a + b * X[i]
    }

    # Prior models for a, b, s




}"

$Y_i \sim N(m_i, s^2)$ for $i$ from 1 to 507
$m_i = a + b X_i$
- NOTE: use <- not ~

DEFINE the regression model

weight_model <- "model{
    # Likelihood model for Y[i]
    for(i in 1:length(Y)) {
      Y[i] ~ dnorm(m[i], s^(-2))
      m[i] <- a + b * X[i]
    }

    # Prior models for a, b, s
    a ~ dnorm(0, 200^(-2))
    b ~ dnorm(1, 0.5^(-2))
    s ~ dunif(0, 20)

}"

$Y_i \sim N(m_i, s^2)$ for $i$ from 1 to 507
$m_i = a + b X_i$
- NOTE: use <- not ~
$a \sim N(0, 200^2)$
$b \sim N(1, 0.5^2)$
$s \sim \text{Unif}(0, 20)$

COMPILE the regression model

# COMPILE the model    
weight_jags <- jags.model(textConnection(weight_model), 
    data = list(X = bdims$hgt, Y = bdims$wgt),
    inits = list(.RNG.name = "base::Wichmann-Hill", .RNG.seed = 2018))

dim(bdims)

507  25

head(bdims$hgt)

head(bdims$wgt)

174.0 175.3 193.5 186.5 187.2 181.5

65.6 71.8 80.7 72.6 78.8 74.8

SIMULATE the regression model

# COMPILE the model    
weight_jags <- jags.model(textConnection(weight_model), 
    data = list(X = bdims$hgt, Y = bdims$wgt),
    inits = list(.RNG.name = "base::Wichmann-Hill", 
                 .RNG.seed = 2018))

# SIMULATE the posterior
weight_sim <- coda.samples(model = weight_jags, 
    variable.names = c("a", "b", "s"), 
    n.iter = 10000)

Addressing Markov chain instability

Standardize the height predictor (subtract the mean and divide by the standard deviation)
Increase chain length

Posterior insights

Let's practice!

Bayesian Modeling with RJAGS