4

I am looking at logistic regression output that shows Z test values. When can I use these to estimate the significance of coefficients, and when do I need to use a Wald chi-squared test?

1 Answers1

7

They are almost certainly the same thing. The "Z test values" in your logistic regression output are probably Wald statistics, and the Wald chi-square statistic is probably just $Z^2$. This means that the Wald chi-squared test is the same as a two-sided test using the corresponding $Z$-value.

Gordon Smyth
  • 12,807
  • 1
    So just to be clear - if I need the Wald's Chi-Square value, all I need to do is take a square of the Z statistic from the output? – Indrajit Feb 05 '18 at 08:15