In goldingn/greta: Simple and Scalable Statistical Modelling in R

Eight schools

Eight schools estimates the effect of coaching programs in eight schools. The data are 8 measurements of coaching effects along with their standard errors.

See also the Stan example for details.

data

y <- c(28,  8, -3,  7, -1,  1, 18, 12)
sigma_y <- c(15, 10, 16, 11,  9, 11, 10, 18)
N  <- length(y)

greta code

sigma_eta <- inverse_gamma(1, 1)
eta <- normal(0, sigma_eta, dim=N)

mu_theta <- normal(0, 100)
xi <- normal(0, 5)
theta <- mu_theta + xi * eta

distribution(y) <- normal(theta, sigma_y)

Stan code

cat(readLines('https://raw.githubusercontent.com/stan-dev/example-models/master/ARM/Ch.19/schools.stan'), sep = '\n')

goldingn/greta documentation built on May 24, 2021, 11 a.m.