I am doing conditional logistic regression in R using clogit from survival. Lets say that Y takes values fail/succes. My variable of interest, X, is a continuous variable with 25th percentile = 0.152 and 75th percentile = 0.396. Its coefficient in the regression is 1.0786. I want to know how to interpret this.
To my understanding, in "normal" logistic regression, e.g., not fixed effects but just cross-sectional, the interpretation would be that a one-unit change in X increases the odds of success with 194% (exp(1.0786) = 2.94056). Moving from the 25th to the 75th percentile would increase the odds of sucess with 30% ( exp((0.396-0.152)*1.0786) = 1.301059).
How well does this interpretation translate to the conditional case? Can I say that a 1 unit change in X increases odds of succes with 194% within groups? Also, is it possible to compute predicted probabilities based on my conditional logistic regression, i.e., probability of succes when 1) X = 25th percentile and 2) X = 75th percentile?
