Premium
Using sparse categorical principal components to estimate asset indices: new methods with an application to rural southeast asia
Author(s) -
Merola Giovanni Maria,
Baulch Bob
Publication year - 2019
Publication title -
review of development economics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.531
H-Index - 50
eISSN - 1467-9361
pISSN - 1363-6669
DOI - 10.1111/rode.12568
Subject(s) - principal component analysis , econometrics , ranking (information retrieval) , categorical variable , asset (computer security) , per capita , economics , estimation , per capita income , statistics , mathematics , computer science , population , demography , computer security , management , machine learning , sociology
Asset indices have been used since the late 1990s to measure wealth in developing countries. We extend the standard methodology for estimating asset indices using principal component analysis in two ways: by introducing constraints that force the indices to have increasing value as the number of assets owned increases, and by estimating sparse indices with a few key assets. This is achieved by combining categorical and sparse principal component analysis. We also apply this methodology to the estimation of per capita level asset indices. Using household survey data from northwest Vietnam and northeast Laos, we show that the resulting asset indices improve the prediction and ranking of income both at household and per capita level.