Example Model and Data
The following is a simple, complete Stan program for a Bernoulli model of binary data.1 The model assumes the binary observed data y[1],...,y[N]
are i.i.d. with Bernoulli chance-of-success theta
.
data {
int<lower=0> N;
array[N] int<lower=0, upper=1> y;
} parameters {
real<lower=0, upper=1> theta;
} model {
1, 1); // uniform prior on interval 0,1
theta ~ beta(
y ~ bernoulli(theta); }
The input data file contains definitions for the two variables N
and y
which are specified in the data block of program bernoulli.stan
(above).
A data set of N=10
observations is included in the example Bernoulli model directory in both JSON notation and Rdump data format where 8 out of 10 trials had outcome 0
(failure) and 2 trials had outcome 1
(success). In JSON, this data is:
{
"N" : 10,
"y" : [0,1,0,0,0,0,0,0,0,1]
}
Footnotes
The model is available with the CmdStan distribution at the path
<cmdstan-home>/examples/bernoulli/bernoulli.stan
.↩︎