Calculate t-stat from stan()

Modeling
1

Hi everyone,
I would like to calculate the t-stat for the estimates obtained through stan() function. Parameter estimates of normal distribution function with true mu=0 and true sigma=1 are given below. I would appreciate any help.

Inference for Stan model:
4 chains, each with iter=1000; warmup=250; thin=1;
post-warmup draws per chain=750, total post-warmup draws=3000.

         mean  se_mean    sd   2.5%   25%    50%    75%   97.5%   n_eff  Rhat 
mu       0.41     0.01  0.36  -0.31  0.19   0.42   0.65    1.11    2201     1
sigma    1.24     0.01  0.28   0.83  1.04   1.20   1.40    1.93    1923     1
lp__    -8.01     0.03  1.02 -10.77 -8.36  -7.70  -7.28   -6.99    1359     1

Samples were drawn using NUTS(diag_e). For each parameter, n_eff is a crude measure of effective sample size, and Rhat is the potential scale reduction factor on split chains (at convergence, Rhat=1).

Thanks!

2

Hm, a t-stat is generally not something you do in the Bayes world, so it sounds like you might benefit from this intro vid: https://www.youtube.com/watch?v=wMTUhMBSFLM&list=PLu77iLvsj_GNmWqDdX-26kZ56dCZqkzBO&index=9&t=0s

1 Like