Deriving the funtion of a model

[Patrick Devers]

I have completed running my a-priori models for survival with 4 occasions using a known fates model with individual covariates but I am unsure how to derive the reconstituted function. The model was
S = B(0) + Forest Type + Age + Sex + Age*Sex + time

Forest type, age, and sex all have 2 levels (coded 0, 1).

and the design matrix was as follows:
1 Forest Age Sex Age*Sex 1
1 Forest Age Sex Age*Sex 1
1 Forest Age Sex Age*Sex 1
1 Forest Age Sex Age*Sex 1

I understand to derive the function I use the equation:

logit(theta) = intercept + B(m-m(bar)/SD) + B(m-m(bar)/SD) + B(m-m(bar)/SD) + B(m-m(bar)/SD)

and then transform back from the logit scale.

But my question is; how do I handle the time component? Do I derive the reconsitituted values for each forest, age, and sex combination and multiply by the time specific survival estimate or do I multiple it by the probability of surviving the entire study? My ultimate goal is to be able to graph the estimate time specific survival estimates and standard errors for each forest type, age, and sex class and to report the estimate difference in survival rates. I appreciate any thoughts or suggestions.

thanks.