Hello everyone, I have a data set that contains characteristics of 25,000 patients of 92 different hospitals. I have run a regression to capture the probability these patients will have a complication after a certain operation. Now, I actually want to predict the probability per patient, using the outcome of the regression, but I just certain patients involved in the prediction. So, I want to use the estimated parameters of the full regression, but after that, I only want patients with certain characteristics in my prediction matrix. For example, only patients with BMI >25.
My regressional formula looks as follows: fit = glmer(response ~ sex+bmi+asa+ (1|center), data=data, family=binomial), and my normal prediction formula: pred=fit@X %*% fit@fixef. I hope someone can help me! Greetings, Alex [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.