I want to work on some malnutrition conditions of women and children using demographic and health survey data. I want to regress the body mass index variables on some correlate variables but there are a large number of missing values in body mass index variables. How do I deal with these missing values so that good results can be found, and research looks meaningful? Furthermore I want to use STATA version 11 for the analysis.
If your data are missing at random, consider using multiple imputation. You could consult Rubin and Little on the Statistical Analysis of missing data.
If your missing observations are not random then your estimation results can be biased. You should first analyse the nature of the missing data. You can also consider to use Heckman selection model if you have a good candidate for the selection IV.
- Badges
- Science topic
- Similar topics
- Mathematics