Search code examples
rr-markdownstanrstan

How to use stan in rmarkdown


I would like to get the estimated coefficients of a model using rstan in an rnotebook

I have the following stan chunk:

```{stan output.var="rats"}
data {
  int<lower=0> N;
  int<lower=0> T;
  real x[T];
  real y[N,T];
  real xbar;
}
parameters {
  real alpha[N];
  real beta[N];

  real mu_alpha;
  real mu_beta;          // beta.c in original bugs model

  real<lower=0> sigmasq_y;
  real<lower=0> sigmasq_alpha;
  real<lower=0> sigmasq_beta;
}
transformed parameters {
  real<lower=0> sigma_y;       // sigma in original bugs model
  real<lower=0> sigma_alpha;
  real<lower=0> sigma_beta;

  sigma_y = sqrt(sigmasq_y);
  sigma_alpha = sqrt(sigmasq_alpha);
  sigma_beta =  sqrt(sigmasq_beta);
}
model {
  mu_alpha ~ normal(0, 100);
  mu_beta ~ normal(0, 100);
  sigmasq_y ~ inv_gamma(0.001, 0.001);
  sigmasq_alpha ~ inv_gamma(0.001, 0.001);
  sigmasq_beta ~ inv_gamma(0.001, 0.001);
  alpha ~ normal(mu_alpha, sigma_alpha); // vectorized
  beta ~ normal(mu_beta, sigma_beta);  // vectorized
  for (n in 1:N)
    for (t in 1:T) 
      y[n,t] ~ normal(alpha[n] + beta[n] * (x[t] - xbar), sigma_y);

}
generated quantities {
  real alpha0;
  alpha0 = mu_alpha - xbar * mu_beta;
}
```

I also have the following data

```{r}
df <- read_delim("https://raw.githubusercontent.com/wiki/stan-dev/rstan/rats.txt",delim = " ")

y <- as.matrix(df)
x <- c(8,15,22,29,36)
xbar <- mean(x)
N <- nrow(y)
T <- ncol(y)
```

The documentation on github shows rats_fit <- stan(file = 'https://raw.githubusercontent.com/stan-dev/example-models/master/bugs_examples/vol1/rats/rats.stan'), but since I am using a chunk I don't have a file to refer to.

I have tried stan(rats), summary(rats), print(rats), but none of these seem to work.


Solution

  • The first RMarkdown chunk calls rats <- rstan::stan_model(model_code=the_text) behind the scenes, so in order to sample from that posterior distribution you need to ultimately do rats_fit <- sampling(rats, data = list()), whose remaining arguments are pretty much the same as for stan. But you do have to call library(rstan) before all that.