Premium
A latent‐class regression approach for the analysis of recurrent choice data
Author(s) -
Böckenholt Ulf
Publication year - 1993
Publication title -
british journal of mathematical and statistical psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.157
H-Index - 51
eISSN - 2044-8317
pISSN - 0007-1102
DOI - 10.1111/j.2044-8317.1993.tb01004.x
Subject(s) - latent class model , latent variable , multivariate statistics , regression analysis , representation (politics) , interpretation (philosophy) , regression , statistics , class (philosophy) , factor regression model , computer science , set (abstract data type) , data set , population , mathematics , latent variable model , econometrics , polynomial regression , artificial intelligence , proper linear model , demography , sociology , politics , political science , law , programming language
This paper introduces a mixture regression model for the analysis and interpretation of multivariate count data in a heterogeneous population. The model is derived by specifying a regression relationship at the individual level between the choice data and a set of concomitant variables. A latent class representation is introduced to describe variations in the regression weights among individuals and to obtain a representation of the aggregate choice behaviour. An easy‐to‐implement EM algorithm is presented for parameter estimation.