[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: weighted estimates for a multiply imputed variable?

From   Michelle L Stransky <>
Subject   st: weighted estimates for a multiply imputed variable?
Date   Tue, 18 Aug 2009 12:11:05 -0400


I've been using the mim: svy command to deal with a multiply imputed dataset with a complex sample design. I've successfully run my logistic regressions and significance tests. However, I'm having trouble calculating a weighted frequency for the multiply imputed variable itself. Five imputations of income were developed for the dataset by Natl Ctr for Health Stats. I'm trying to calculate estimates of the number of children with Down Syndrome who fall into each family income level.

I'm using Stata9 and I cannot figure out which of the mim: svy commands (proportion, mean, ratio, etc.) will yield weighted frequencies (I've already calculated weighted percentages and counts for all of the other variables). I assume that this can be done because I have an article that calculated this weighted frequency on this same dataset using SUDAAN.

I'm pretty sure I'm missing something fairly obvious and would appreciate someone pointing this out to me.



*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index