| Re: Merging data files in Stata [message #2227 is a reply to message #2218] |
Fri, 30 May 2014 09:53  |
ahmed89o
Messages: 26
Registered: August 2013
Location: Germany
|
Member |
|
|
Many thanks for this. I have some comments here.
I am using Egypt DHS from 1995 to 2008 so if antenatal care is only for youngest child born in the last five years. How am getting the same number of observations for m2a and m3a? Am assuming that m2a must be less in such case but they are not?
"the subgroup of youngest children usually has a better chance of surviving than the other children born in the past five years (the youngest child tends to have a longer preceding birth interval)." can you please clarify a bit more.
LPM vs probit, my reference is Wooldridge (introductory to economitrics) he still recommending with some adjustments like Weight least square. I economics student not public health, as you may know that we are more into marginal effects than odds ratio. I found it complicated to calculate marginal effects for interaction terms either in stata or SPSS (I think SPSS cannot do it all). If you can advise how to get marginal effect correctly in stata I will be grateful.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
