Closed Azhag closed 12 years ago
Continuing from Issue #20, check if the FI estimated from the precision of the samples is a good estimator for the FI estimated from other means.
Seems that there is a finite sample effect. But good direction.
We actually need to compare the FI from the posterior, from samples from theta_i and the precision from sampled recalls. See #27, #28, #29
Continuing from Issue #20, check if the FI estimated from the precision of the samples is a good estimator for the FI estimated from other means.