The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Infant and Under-five Mortality in Myanmar
Infant and Under-five Mortality in Myanmar [message #15042] Mon, 28 May 2018 01:00 Go to next message
maimawite@gmail.com is currently offline  maimawite@gmail.com
Messages: 3
Registered: January 2018
Location: Thailand
Member
Dear Sir,
Let me ask questions on using weight on DHS data. As per guidelines and video clip from the webside, I do understand that using of weight on analysis of the data.

I am working on factors associated with infant and under-five mortality in Myanmar and I would like to do analysis by using of methodology called logit model and magnitude of impact by Maximum likelihood method and not with ODD ratio.

I do not know how to use stata code to get the point.
When I use svy: quietly logit INM X1 X2 ..... and the results showed with logistic regression.

From the forum, I learned alot on my paper and thanks for all of your contribution.

Best wishes,
Lal Hrin Mawi


Lal Hrin Mawi
Re: Infant and Under-five Mortality in Myanmar [message #15061 is a reply to message #15042] Tue, 29 May 2018 13:05 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
Can you please provide more details on what you are trying to accomplish? Perhaps you can share your code so that others in the user community may be able to assist.
You can also search the forum for posts on infant mortality. Other useful tools: The Guide to DHS Statistics https:// dhsprogram.com/publications/publication-dhsg1-dhs-questionna ires-and-manuals.cfm, The Standard Recode Manual https:// dhsprogram.com/publications/publication-DHSG4-DHS-Questionna ires-and-Manuals.cfm a series of YouTube videos: https://blog.dhsprogram.com/dhsdataintro/ and for help with Stata: http://www.cpc.unc.edu/research/tools/data_analysis/statatutorial
Thank you!
Re: Infant and Under-five Mortality in Myanmar [message #15069 is a reply to message #15061] Wed, 30 May 2018 00:54 Go to previous message
maimawite@gmail.com is currently offline  maimawite@gmail.com
Messages: 3
Registered: January 2018
Location: Thailand
Member
Dear Liz,

Thanks for your quick reply and additional links for further study on statistics. Mostly, the studies were done with odd ratios, ΟR=γ(1)/1-r(1)γ(0)/1-r(0).

In my case, I would like to use the methodologies called ,Maximum likelihood method using in non-liner equation or econometric model.

Please let me share my some work done so far.

***** Factors associated with Infant and Under-five mortality in Myanmar"
replace INM = 1 if b7 <12&!missing(b7)
replace INM = 0 if INM == .
generate CM = .
replace CM = 1 if inrange(b7,12,48)
replace CM=0 if CM ==.
generate U5M = .
replace U5M = 1 if inrange(b7,0,48)
replace U5M = 0 if U5M == .

gen wgt = v005/1000000
gen psu = v021
gen strata = v023
svyset psu [pw = v021], strata(v022)

recode v212 (12/19=0) (20/35=1) (36/43=2), gen(Age)
gen WealthQ1 = 1 if v190 == 1
replace WealthQ1 = 0 if WealthQ1 == .
gen WealthQ2 = 1 if v190 == 2
replace WealthQ2 = 0 if WealthQ2 == .
gen WealthQ3 = 1 if v190 == 3
replace WealthQ3 = 0 if WealthQ3 == .
gen WealthQ4 = 1 if v190 == 4
replace WealthQ4 = 0 if WealthQ4 == .
gen WealthQ5 = 1 if v190 == 5
replace WealthQ5 = 0 if WealthQ5 == .
gen female=1 if b4==2
replace female=0 if b4==1
gen twin=1 if b0==1|b0==2|b0==3
replace twin=0 if b0==0
gen Breastfeeding=1 if m4==93| m4==95
replace Breastfeeding=0 if m4==94
replace Breastfeeding=0 if Breastfeeding==.

Explanatory variables used are Age of mother at birth; wealth qunitles of HH;twin;female;breasefeeding etc

Stata command is as follows:
quietly logit INM Age WealthQ2 WealthQ3 WealthQ4 WealthQ5 feamle twin Breastfeeding
margins, dydx(*)

quietly logit INM i.Age WealthQ2 WealthQ3 WealthQ4 WealthQ5 female twin Breastfeeding

. margins, dydx(*)

Average marginal effects Number of obs = 4815
Model VCE : OIM

Expression : Pr(INM), predict()
dy/dx w.r.t. : 1.Age 2.Age WealthQ2 WealthQ3 WealthQ4 WealthQ5 female twin
Breastfeeding


and I tried to use svy:quietly logit INM Age WealthQ2 WealthQ3 WealthQ4 WealthQ5 feamle twin Breastfeeding;
svy: quietly logit INM i.Age WealthQ2 WealthQ3 WealthQ4 WealthQ5 female twin Breastfeeding
(note: ignoring quietly)
(running logit on estimation sample)

Survey: Logistic regression


The results are shown with logistic regression.

I did not know how to use weight in logit model with marginal effect. Could you kindly provide me stata command used for econometric model.

Thanks in advance for your input.
Best Wishes,
Lal






Lal Hrin Mawi
Previous Topic: Domestic Violence and Infant Mortality- NFHS4
Next Topic: Child mortality indicators in AIS survey
Goto Forum:
  


Current Time: Thu Mar 28 05:01:34 Coordinated Universal Time 2024