Marginal contribution of each variable in logistic regression

[Boriana]

Can I estimate the marginal contribution of each variable in logistic regression by Standardized Estimate, received in stb option?

Can I calculate marginal contribution of variables by other options or formula?

[PaigeMiller · Accepted solution]

I don't think the concept of marginal contribution of a variable in a regression model exists. When your x-variables are correlated with one another, they don't split up the sum of squares of the model uniquely or independently. The effect of a term in the model changes as other terms are added or removed from the model, and so there isn't even a unique effect of each term (exception: if your X-matrix is orthogonal then there is a unique effect of each term).

[Boriana]

Thank you for the answer. Did you know the simple explanation of the Standardized Estimate in the logistic regression model?

[PaigeMiller]

From the documentation at https://documentation.sas.com/?cdcId=pgmmvacdc&cdcVersion=9.4&docsetId=statug&docsetTarget=statug_lo...

[StatDave]

See this note on assessing variable importance. The partial correlation for each variable is available as discussed.

[Boriana]

Yes, I know that Standardized Estimate can be used, but we received question for other estimation on this topics.