Multiple regression models, AIC and PASW


New Member

I would like to calculate AIC values from the different models I have created via stepwise multiple regression in the program PASW (SPSS). I have the mac version. The model fit option for the linear regression does not give me this info. I would greatly appreciate any help here..

Thankyou Kindly



TS Contributor
I know 2 ways

1) Run a mixed model (without using a
random factor) but use Maximum Likelihood Estimation (and not REML) to get the AIC of the model.

2) Simply, calculate AIC = n *log(residual sum of squares/n) + 2 * k, where k is the number of parameters. NOTE: Include the variance as a parameter in counting!!!