T-Test of a Regression Coefficient

From Q
Jump to: navigation, search

Where \hat{\beta} is a coefficient, n is the effective number of observations (note that this is different to the effective sample size where repeated observations exist for each respondent) and se_{\beta} is the standard error of the coefficient, the test statistic is:

t=\frac{\hat{\beta}}{s_{\beta}},

where:

p = 2\Pr(t_{n-k} \ge |t|),
k is the number of parameters, including the intercept (where applicable).
Personal tools
Namespaces

Variants
Actions
Navigation
Categories
Toolbox