Plotting odds ratios at different time points

[Elyse Cottrell]

Hello,

I'm looking to create a graph that plots the odds ratios for a model I'm working on but I'm having a difficult time figuring out how to best do it.

We are looking at the odds ratios of falls by location and sex, and want a visual to show the difference over four time points.

If I run:

logistic FALL location##c.time##sex

I get the table that shows me the overall odds ratios (which I'm reporting) but we also want to see the difference between four time points.

if I run:

logistic FALL location##c.time##sex if time==1
logistic FALL location##c.time##sex if time==2
logistic FALL location##c.time##sex if time==3
logistic FALL location##c.time##sex if time==4

coefplot, drop(_cons) xline(0) eform xtitle(Odds ratio)

I get collinearity errors (which makes sense because I'm putting time in there twice). I'm not quite clear if taking c.time out will change the model (in that I would then be measuring interactions completely different from the way they were measured in the first code, if that makes sense?)

The other issue is the coefplot is only plotting one of the time points and I want all four on the same graph (using the same axis).

Any help is greatly appreciated.

Thank you!