| Re: How to arrive at final wealth index score using PCA [message #31342 is a reply to message #28666] |
Sat, 23 August 2025 10:18   |
|
|
hi, i think in DHS methodology, the wealth index is indeed based on the first principal component (PCA). Using predict in Stata after running PCA will give you the factor scores, which can be used to rank households. Two clarifications:
Yes, taking the first predicted factor score (usually wealthscore1) is sufficient to construct the wealth index.
Households are then ranked and divided into quintiles using survey weights (as you did with xtile).
|
|
|
|
Members
Search
Help
Register
Login
Home
