IBM Support

Can SPSS produce AIC or BIC for logistic regression models?

Troubleshooting


Problem

I'm running a logistic regression model in SPSS Statistics and would like to see AIC and/or BIC values. Can SPSS provide these?

Resolving The Problem

The Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) are available in the NOMREG (Multinomial Logistic Regression in the menus) procedure. In command syntax, specify the IC keyword on the /PRINT subcommand. In the dialog boxes, click on the Statistics button and check the Information criteria check box.

Note that by default the NOMREG procedure prints information criteria based on the full -2 log-likelihood, which includes the binomial or multinomial constant. This constant is often excluded by other programs and procedures, including the LOGISTIC REGRESSION procedure (Binary Logistic Regression in the menus) which print log-likelihood values based only on the kernel of the log-likelihood. To make NOMREG results match those results, in command syntax add the keyword KERNEL to the /PRINT subcommand. See Technote 1476887 for more details.

[{"Product":{"code":"SSLVMB","label":"IBM SPSS Statistics"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"Not Applicable","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Historical Number

71969

Document Information

Modified date:
16 April 2020

UID

swg21478171