Time series analysis - different results for same cross correlation in SPSS?!

#1
Hey everyone,
i have a question about the results I get when I do a cross correlation of two white noise time series with SPSS 21.0. Why do I get two different results depending on how much variables I insert in the window for the cross correlation?

For example: If I insert the variables A,B,C,D,E,F in the window for cross correlation, SPSS shows me a significant cross correlation between A and B. But if I insert only the variables A,B,C, the cross correlation between A and B is no longer significant. Is SPSS controlling for the other variables I insert?

Thanks in advance!
 

Attachments

Dason

Ambassador to the humans
#2
I don't use SPSS but is it possible you have missing values and it's removing rows where you have any missing values for any of the variables
 
#3
I had missing values, but I already had estimated them by linear interpolation before I modeled my white noise residuals and tested for cross correlation.