I am using the PROC LOGISTIC to conduct exact logistic regression. There are two main exposures: A and B. Both are two-level factors. I am interested in the effects of A, B and their interaction on a binary response variable, adjusted by other covariates. So, the logit is: beta0 + beta1*A + beta2*B + beta3*AB + other-covariates-terms.
After reading the documentation, I realized that the estimate of "beta1 + beta2 + beta3" is not available in the exact analysis. Did I understand correctly?