The DHS Program User Forum: Sampling » Fixed effects and svy

Home » Data » Sampling » Fixed effects and svy

Show: Today's Messages :: Show Polls :: Message Navigator

Fixed effects and svy [message #10926]

Fri, 07 October 2016 08:52

shayankhan
Messages: 16
Registered: September 2016
Location: Karachi, Pakistan

Member

I want to run my logit model with fixed effects to control for the unobserved heterogeneity across different districts. So, I use xtset command to set the data according to the districts and then I run the xtlogit command with fixed effects. However, when I use another way to get the same results that is by using svy: logit x z i.y, where i.y represents the districts fixed effect dummies, I get different results. So, I want to know which method is correct. Can I Ignore the svy command while running my fixed effect model?

Report message to a moderator