Two applications of the random coefficient procedure: Correcting for misspecifications in a small area level model and resolving Simpson's paradox

We apply a random-coefficient framework to deal with two problems frequently encountered in applied work. First, we use a real-world relationship to derive a sub-relationship among fewer variables without introducing any specification error to correct misspecifications in a small area level model. S...

Full description

Saved in:
Bibliographic Details
Published in:Economic modelling 2015-02, Vol.45, p.93-98
Main Authors: Swamy, P.A.V.B., Mehta, J.S., Tavlas, G.S., Hall, S.G.
Format: Article
Language:English
Subjects:
Coefficients
Economic models
Economic theory
Error
Error analysis
False model
Mathematical problems
Random variables
Specification error
Studies
Unique coefficient
Unique error term
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We apply a random-coefficient framework to deal with two problems frequently encountered in applied work. First, we use a real-world relationship to derive a sub-relationship among fewer variables without introducing any specification error to correct misspecifications in a small area level model. Second, we then use this framework to resolve Simpson's paradox. We show that this paradox does not arise if a statistical relationship between a pair of variables is derived from the corresponding real-world relationship involving all relevant variables, including the original pair, without introducing a single specification error. •We apply a random-coefficient framework to deal with two problems.•First, we derive a sub-relationship in a small area level model.•The sub-relationship is free of specification errors.•Second, we use this framework to resolve Simpson’s paradox.
ISSN:0264-9993
1873-6122
DOI:10.1016/j.econmod.2014.10.053