Bayesian semi-parametric analysis of Poisson change-point regression models

Application to policy-making in Cali, Colombia

Taeyoung Park, Robert T. Krafty, Alvaro I. Sánchez

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

A Poisson regression model with an offset assumes a constant baseline rate after accounting for measured covariates, which may lead to biased estimates of coefficients in an inhomogeneous Poisson process. To correctly estimate the effect of time-dependent covariates, we propose a Poisson change-point regression model with an offset that allows a time-varying baseline rate. When the non-constant pattern of a log baseline rate is modeled with a non-parametric step function, the resulting semi-parametric model involves a model component of varying dimensions and thus requires a sophisticated varying-dimensional inference to obtain the correct estimates of model parameters of a fixed dimension. To fit the proposed varying-dimensional model, we devise a state-of-the-art Markov chain Monte Carlo-type algorithm based on partial collapse. The proposed model and methods are used to investigate the association between the daily homicide rates in Cali, Colombia, and the policies that restrict the hours during which the legal sale of alcoholic beverages is permitted. While simultaneously identifying the latent changes in the baseline homicide rate which correspond to the incidence of sociopolitical events, we explore the effect of policies governing the sale of alcohol on homicide rates and seek a policy that balances the economic and cultural dependencies on alcohol sales to the health of the public.

Original languageEnglish
Pages (from-to)2285-2298
Number of pages14
JournalJournal of Applied Statistics
Volume39
Issue number10
DOIs
Publication statusPublished - 2012 Oct 1

Fingerprint

Change-point Model
Parametric Analysis
Baseline
Regression Model
Siméon Denis Poisson
Alcohol
Inhomogeneous Poisson Process
Estimate
Time-dependent Covariates
Poisson Regression
Step function
Semiparametric Model
Poisson Model
Component Model
Markov Chain Monte Carlo
Biased
Covariates
Incidence
Time-varying
Health

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{b8a2b1a2e6e24cb6a83f27d9426aea09,
title = "Bayesian semi-parametric analysis of Poisson change-point regression models: Application to policy-making in Cali, Colombia",
abstract = "A Poisson regression model with an offset assumes a constant baseline rate after accounting for measured covariates, which may lead to biased estimates of coefficients in an inhomogeneous Poisson process. To correctly estimate the effect of time-dependent covariates, we propose a Poisson change-point regression model with an offset that allows a time-varying baseline rate. When the non-constant pattern of a log baseline rate is modeled with a non-parametric step function, the resulting semi-parametric model involves a model component of varying dimensions and thus requires a sophisticated varying-dimensional inference to obtain the correct estimates of model parameters of a fixed dimension. To fit the proposed varying-dimensional model, we devise a state-of-the-art Markov chain Monte Carlo-type algorithm based on partial collapse. The proposed model and methods are used to investigate the association between the daily homicide rates in Cali, Colombia, and the policies that restrict the hours during which the legal sale of alcoholic beverages is permitted. While simultaneously identifying the latent changes in the baseline homicide rate which correspond to the incidence of sociopolitical events, we explore the effect of policies governing the sale of alcohol on homicide rates and seek a policy that balances the economic and cultural dependencies on alcohol sales to the health of the public.",
author = "Taeyoung Park and Krafty, {Robert T.} and S{\'a}nchez, {Alvaro I.}",
year = "2012",
month = "10",
day = "1",
doi = "10.1080/02664763.2012.709227",
language = "English",
volume = "39",
pages = "2285--2298",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",
number = "10",

}

Bayesian semi-parametric analysis of Poisson change-point regression models : Application to policy-making in Cali, Colombia. / Park, Taeyoung; Krafty, Robert T.; Sánchez, Alvaro I.

In: Journal of Applied Statistics, Vol. 39, No. 10, 01.10.2012, p. 2285-2298.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Bayesian semi-parametric analysis of Poisson change-point regression models

T2 - Application to policy-making in Cali, Colombia

AU - Park, Taeyoung

AU - Krafty, Robert T.

AU - Sánchez, Alvaro I.

PY - 2012/10/1

Y1 - 2012/10/1

N2 - A Poisson regression model with an offset assumes a constant baseline rate after accounting for measured covariates, which may lead to biased estimates of coefficients in an inhomogeneous Poisson process. To correctly estimate the effect of time-dependent covariates, we propose a Poisson change-point regression model with an offset that allows a time-varying baseline rate. When the non-constant pattern of a log baseline rate is modeled with a non-parametric step function, the resulting semi-parametric model involves a model component of varying dimensions and thus requires a sophisticated varying-dimensional inference to obtain the correct estimates of model parameters of a fixed dimension. To fit the proposed varying-dimensional model, we devise a state-of-the-art Markov chain Monte Carlo-type algorithm based on partial collapse. The proposed model and methods are used to investigate the association between the daily homicide rates in Cali, Colombia, and the policies that restrict the hours during which the legal sale of alcoholic beverages is permitted. While simultaneously identifying the latent changes in the baseline homicide rate which correspond to the incidence of sociopolitical events, we explore the effect of policies governing the sale of alcohol on homicide rates and seek a policy that balances the economic and cultural dependencies on alcohol sales to the health of the public.

AB - A Poisson regression model with an offset assumes a constant baseline rate after accounting for measured covariates, which may lead to biased estimates of coefficients in an inhomogeneous Poisson process. To correctly estimate the effect of time-dependent covariates, we propose a Poisson change-point regression model with an offset that allows a time-varying baseline rate. When the non-constant pattern of a log baseline rate is modeled with a non-parametric step function, the resulting semi-parametric model involves a model component of varying dimensions and thus requires a sophisticated varying-dimensional inference to obtain the correct estimates of model parameters of a fixed dimension. To fit the proposed varying-dimensional model, we devise a state-of-the-art Markov chain Monte Carlo-type algorithm based on partial collapse. The proposed model and methods are used to investigate the association between the daily homicide rates in Cali, Colombia, and the policies that restrict the hours during which the legal sale of alcoholic beverages is permitted. While simultaneously identifying the latent changes in the baseline homicide rate which correspond to the incidence of sociopolitical events, we explore the effect of policies governing the sale of alcohol on homicide rates and seek a policy that balances the economic and cultural dependencies on alcohol sales to the health of the public.

UR - http://www.scopus.com/inward/record.url?scp=84865818325&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865818325&partnerID=8YFLogxK

U2 - 10.1080/02664763.2012.709227

DO - 10.1080/02664763.2012.709227

M3 - Article

VL - 39

SP - 2285

EP - 2298

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

IS - 10

ER -