Why is t-statistics of glmfit based on standard errors instead of dispersion

1 view (last 30 days)
Fan
Fan on 1 Sep 2011
t-statistics (F-statistics) could be based on the reduction of the standard error or the likelihood (i.e. dispersion). Since glmfit solves regression by maximizing likelihood, why its t-statistics and p-values are based on the standard error rather than dispersion?

Sign in to answer this question.

Answers (0)

Categories

Find more on Regression in Help Center and File Exchange

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!