Panel data regression time fixed & firm fixed

Hi all,

After desparately searching the web and the library I hope one of you will be able to help me with my spss (no 19) problem.
My data set consists out of two time periods (recession vs boom) each including 4 observations per bank for 4600 banks. Furthermore the banks are organized into two groups (basel II vs no basel II).
The theme of my research is the procyclical influence of Basel II.

I operationalized this in a regression taking the form of ROA= a + B1*dummybaselIIvsnobasel+B2*dummyrecessionboom+B3*interactioneffectdummy1&dummy2+ Control variables+ e.
A posotive B3 then would mean a procyclical influence of Basel II (dummy1=1 if Basel II and dummy2=1 if recession).
I already performed OLS regression but due to the firm or time interaction I feel that these results might not be reliable. I would like to correct for that but I don't know how. Could anybody help me with this? If you need more information or anything please let me know.
Many thanks in advance