Re: likelihood model for missing SEX data

From: Anthe Zandvliet Apaza@slz.nl Subject: Re: [NMusers] likelihood model for missing SEX data Date: Fri, 07 Jul 2006 17:23:52 +0200 Jakob, You might be right about the upward bias. I am planning to perform simulations to evaluate this. I also agree that any other covariates should be taken into account if possible. Yet I wonder what method would be best if no other covariates are available and if the effect of SEX on CL is substantial. I am interested in your suggestion to use individual probabilities and a linear covariate model. Kristin, Is your script to calculate IOFVs available online? Best regards, Anthe