Mixture Models: Identifying Consumption Classes in Post-liberalization India

Sudeshna Maitra

Chapter Chapter 5 in Applied Econometric Analysis Using Cross Section and Panel Data, 2023, pp 135-166 from Springer

Abstract: Abstract A mixture model is a probabilistic model that allows us to make inferences about the characteristics of sub-populations from observations on the overall population, without any information about the membership of individuals in the sub-populations or even the number of sub-populations. In this chapter, I present the theory of mixture models, and an application in which I identify consumption classes in urban India in 1999–00 (NSS). Suppose there are three sub-populations—a lower, a middle, and an upper consumption class—determined by the total number of different durables owned by households. I construct a three-component (or three-class) mixture model of household durable ownership, which is assumed to be distributed binomially by class. I then demonstrate the use of the Expectation Maximization (EM) algorithm to estimate the size of and mean durables owned by each class, as well as the probability that a household with a given number of durables belongs to a given class. Finally, I show how to assign households to classes using the mixture estimates, which allows further investigation of class-specific characteristics.

Keywords: Consumption class; Binomial mixture model; EM algorithm (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:conchp:978-981-99-4902-1_5

Ordering information: This item can be ordered from
http://www.springer.com/9789819949021

DOI: 10.1007/978-981-99-4902-1_5

Access Statistics for this chapter

More chapters in Contributions to Economics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().