Z-values in Logistic Regression stem from a Wald test?

W

Wicked

Am I correct in thinking that Z-values, and their corresponding P-values, I find in logistic regression stem from a Wald test?
 

Statistical Steven

Statistician
Leader
Super Moderator
Am I correct in thinking that Z-values, and their corresponding P-values, I find in logistic regression stem from a Wald test?

What software do you use? The Wald test is a Chi-Square test. I assume you meant chi-square values and not z-values. Can you clarify for me?
 
W

Wicked

I'm using Minitab and Binary Logistic Regression.
I get a Z-value next to the P-value for every explanatory variable.
 

Statistical Steven

Statistician
Leader
Super Moderator
I'm using Minitab and Binary Logistic Regression.
I get a Z-value next to the P-value for every explanatory variable.

That table with Z-values is a two-sample test (basically a t-test) but using the standard normal to calculate the p-value.
 

Miner

Forum Moderator
Leader
Admin
Am I correct in thinking that Z-values, and their corresponding P-values, I find in logistic regression stem from a Wald test?
Yes, the Z-values used in Logistic regression do come from the Wald test.

While Minitab does not state this, you can look up the formula for Z in the Logistic regression help menu (Z = coefficient/SE) and compare it to the formula for the Wald test. They are identical. In addition, Minitab does include a reference to a book on the Wald test.
 
W

Wicked

Also, I'm trying to interpret the result from the Pearson Goodness-of-fit test in the same regression. It has a P-value of 0,476, and I'm a tad lost
 

Statistical Steven

Statistician
Leader
Super Moderator
Also, I'm trying to interpret the result from the Pearson Goodness-of-fit test in the same regression. It has a P-value of 0,476, and I'm a tad lost

From Minitab Help

"Goodness-of-Fit Tests - displays Pearson, deviance, and Hosmer-Lemeshow goodness-of-fit tests. In addition, two Brown tests-general alternative and symmetric alternative-are displayed because you have chosen the logit link function and the selected option in the Results subdialog box. The goodness-of-fit tests, with p-values ranging from 0.312 to 0.724, indicate that there is insufficient evidence to claim that the model does not fit the data adequately. If the p-value is less than your accepted a-level, the test would reject the null hypothesis of an adequate fit."

Therefore your p-value is indicative of a good model fit.
 
Top Bottom