Multicollinearity and SigmaStat


New Member
Hello all,
First, thank you for taking the time to read and answer my question!

I am doing a multiple logistic regression analysis in Sigma stat. My dependent variable is dichotomous and my independent variable is categorical ranging from 0-5 however I set up my independent variable as a dichotomous variable (i.e. for category 3, 0=1,2,4,5, 1=3, for category 4, 0=1,2,3,5, 1=4 etc..).
When I try to analyze my data, it seems that Sigma stat just removes one of my independent variables and runs the analysis. I believe it's because of the way I set up my independent variable (since any one of the independent variables is mutually exclusive of the others).

Is my analysis still valid? And if not, how can I circumvent this problem?
Thanks again!


Ambassador to the humans
It's probably doing this because (at least the way I'm understanding what you're saying) for any of the variables we can figure out it's value from the other 4. For example V5 = 1 - V1 - V2 - V3 - V4. To make the model full rank it probably just removes one of the variables. You'll have all the information you need from the reduced model but you just have to make sure you know how you're interpreting it.

Note: I've never used Sigma Stat but that's how most packages deal with that type of data.