# Beta-Binomial
model{
 y ~ dbin(theta,10)
 theta ~ dbeta(1,1) 
}
list(y=7)
list(theta=.5)


#

model{ 
 y ~ dbin(theta, n)  # Model the data
 ytilde ~ dbin(theta, m) # Prediction of future binomial
 theta ~ dbeta(a, b) # The prior
 prob <- step(ytilde - 20) # Pred prob that ytilde >= 20
} 
list(n = 100, m = 100, y = 10, a = 1, b = 1) # The data
list(theta = 0.5, ytilde =10) # Starting/initial values


#

model{ 
 y1 ~ dbin(theta1, n1) # Modeling the data as
 y2 ~ dbin(theta2, n2) # independent binomials
 theta1 ~ dbeta(1, 1)  # Specifying the priors
 theta2 ~ dbeta(1, 1) 
 prob1 <- step(theta1-theta2) # Pr(theta1 >= theta2|data)
 y1tilde ~ dbin(theta1, m1) # Get predictive densities
 y2tilde ~ dbin(theta2, m2)
 prob2 <- step(y1tilde - y2tilde -11) 
}
list(y1=25, y2=10, n1=100, n2=100, m1=100, m2=100) # The data
list(theta1=.5, theta2=.5, y1tilde=20, y2tilde=20) #Starting values