The DHS Program User Forum: Weighting data » Interpretation of Rescaled household level weights for India-NFHS4

Home » Data » Weighting data » Interpretation of Rescaled household level weights for India-NFHS4 (rescaling household level weights for multinomial logit regression and interpretation)

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

Re: Interpretation of Rescaled household level weights for India-NFHS4 [message #18686 is a reply to message #18685]

Tue, 28 January 2020 16:13

preshit
Messages: 13
Registered: March 2018
Location: Tucson, AZ, USA

Member

Ms. Bridgette and Dr. Pullum,

Thank you so much for your useful suggestions. My apologies for the delayed response. I tried all the suggestions given by Stata Tech Support and you but issue did not resolve. I tried excluding some states as well having lowest values of the weight variable but issue persists with the urban subsample. So, I have changed my model specification as follows:

svy : mlogit Y X1##(i.X2 i.X3 i.respondent_residence)) i.X4 i.X5 i.State Fixed Effect
mlogit,rrr

then I interpret all three interaction terms for all my outcomes. The weight variable I use is wt=hv005/1000000. I was reading somewhere that interaction model is more useful than stratified models, so I am OK with this approach. I am getting the output and standard errors are not inflated. Please let me know if you have found solution for urban/rural stratified analysis.

Regards
Preshit

Report message to a moderator

[Message index]

		Interpretation of Rescaled household level weights for India-NFHS4 By: preshit on Mon, 20 January 2020 11:28
		Re: Interpretation of Rescaled household level weights for India-NFHS4 By: Bridgette-DHS on Fri, 24 January 2020 10:00
		Re: Interpretation of Rescaled household level weights for India-NFHS4 By: Bridgette-DHS on Tue, 28 January 2020 16:01
		Re: Interpretation of Rescaled household level weights for India-NFHS4 By: preshit on Tue, 28 January 2020 16:13
		Re: Interpretation of Rescaled household level weights for India-NFHS4 By: Bridgette-DHS on Tue, 28 January 2020 18:35
		Re: Interpretation of Rescaled household level weights for India-NFHS4 By: preshit on Tue, 28 January 2020 18:53

Previous Topic:	Using weights in regression analysis
Next Topic:	Weighting and Multilevel Logistic Regression

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Aug 17 19:16:14 Coordinated Universal Time 2024