Dear Statalisters,
I have a peculiar question on regression models. I feel like I know the answer, but can’t quite get my brain around it.
I’m running two regression models on a data consisting of hormone measurements on pregnant women measured at 3 points in time. The women are divided into three mutually exclusive groups: normal, preterm birth outcome (ptb) and infection. I made two dummy variables: hvptb: which would be 0 if normal women and 1 if ptb, and hvinf: which would be 0 if normal women and 1 if the women had infection problems. My first regression model is
Log(crh)= b0 + b1*time + b2*hvptb + b3*(time*hvptb).
