Computational aspects concerning a model for clustered binary panel data are analysed. The model is based on the representation of the behavior of a subject (individual panel member) in a given cluster by means of a latent process that is decomposed into a cluster-specific component, which follows a first-order Markov chain, and an individual-specific component, which is timeinvariant and is represented by a discrete random variable. In particular, an algorithm for computing the joint distribution of the response variables is introduced. The algorithm may be used even in the presence of a large number of subjects in the same cluster. Also an Expectation- Maximization (EM) scheme for the maximum likelihood estimation of the model is described showing how the Fisher information matrix can be estimated on the basis of the numerical derivative of the score vector. The estimate of this matrix is used to compute standard errors for the parameter estimates and to check the identifiability of the model and the convergence of the EM algorithm. The approach is illustrated by means of an application to a dataset concerning Italian employees’ illness benefits.

Nigro, V., Bartolucci, F. (2007). Maximum likelihood estimation of an extended latent Markov model for clustered binary panel data.

Maximum likelihood estimation of an extended latent Markov model for clustered binary panel data

2007-03-01

Abstract

Computational aspects concerning a model for clustered binary panel data are analysed. The model is based on the representation of the behavior of a subject (individual panel member) in a given cluster by means of a latent process that is decomposed into a cluster-specific component, which follows a first-order Markov chain, and an individual-specific component, which is timeinvariant and is represented by a discrete random variable. In particular, an algorithm for computing the joint distribution of the response variables is introduced. The algorithm may be used even in the presence of a large number of subjects in the same cluster. Also an Expectation- Maximization (EM) scheme for the maximum likelihood estimation of the model is described showing how the Fisher information matrix can be estimated on the basis of the numerical derivative of the score vector. The estimate of this matrix is used to compute standard errors for the parameter estimates and to check the identifiability of the model and the convergence of the EM algorithm. The approach is illustrated by means of an application to a dataset concerning Italian employees’ illness benefits.
mar-2007
Settore SECS-P/05 - ECONOMETRIA
en
EM algorithm
finite mixture models
heterogeneity
latent class model
State dependence
Nigro, V., Bartolucci, F. (2007). Maximum likelihood estimation of an extended latent Markov model for clustered binary panel data.
Nigro, V; Bartolucci, F
Altro
File in questo prodotto:
File Dimensione Formato  
SSRN-id967378.pdf

accesso aperto

Dimensione 398.92 kB
Formato Adobe PDF
398.92 kB Adobe PDF Visualizza/Apri
SSRN-id967378.pdf

accesso aperto

Dimensione 398.92 kB
Formato Adobe PDF
398.92 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2108/351
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact