Ole Melkevik posted on Tuesday, February 21, 2012 - 12:40 pm
I have been struggeling to run a fairly simple monte carlo simulation.
My aim is to investigate the extent to which non-random missingness influence the intercept and estimates of association between a continous predictor and a dichotomous outcome. I basically want to run a simple logistic regression model where I manipulate the extent to which missingness is associated with the the predictor and the dichotomous dependent outcome.
I would be grateful for any feedback or perhaps an example of a basic syntax to work from.