7.00.02 - Wald Test - Aster Analytics

Teradata Aster® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Release Date
September 2017
Content Type
Programming Reference
User Guide
Publication ID
B700-1022-700K
Language
English (United States)

Significance tests can be performed for individual regression coefficients (that is, H0 : βj = 0) by computing the Wald statistics, which are similar to the partial t-statistics from classical regression:



Under the null hypothesis that βj = 0, the Wald test statistic wj follows approximately a standard normal distribution (and its square is approximately a chi-square on one-degree of freedom).

This quantity is computed by the GLM function as the Wald Test, as well as the corresponding 'p_value'. It is in the output table, and also displayed on the screen.