Assessing influence for pharmaceutical data in zero‐inflated generalized Poisson mixed models

For clustered count data with excess zeros where the observations are either over-dispersed or under-dispersed, the zero-inflated generalized Poisson mixed (ZIGPM) regression model may be appropriate, in which the baseline discrete distribution is a generalized Poisson distribution, which is a natural extension of standard Poisson distribution. Motivated by one data set drawn from a pharmaceutical study, influence diagnostics for ZIGPM models based on case-deletion and local influence analysis are developed in this work. The one-step approximations of the estimates under case-deletion model and some case-deletion measures are given. Meanwhile, local influence measures are obtained under various perturbations of the observed data or model assumptions. Results from a pharmaceutical study illustrate the usefulness of the diagnostic statistics.

[1]  Ram C. Tripathi,et al.  Score Test for Zero Inflated Generalized Poisson Regression Model , 2005 .

[2]  Ronald Christensen,et al.  Case-deletion diagnostics for mixed models , 1992 .

[3]  Y. Cheung,et al.  Zero‐inflated models for regression analysis of count data: a study of growth and development , 2002, Statistics in medicine.

[4]  D. Hall.  Zero‐Inflated Poisson and Binomial Regression with Random Effects: A Case Study , 2000, Biometrics.

[5]  Andy H. Lee,et al.  A score test for zero‐inflation in correlated count data , 2006, Statistics in medicine.

[6]  W. Fung,et al.  Sensitivity of score tests for zero‐inflation in count data , 2004, Statistics in medicine.

[7]  C. Mcgilchrist.  Estimation in Generalized Mixed Models , 1994 .

[8]  F. Famoye,et al.  Generalized poisson regression model , 1992 .

[9]  R. Cook.  Detection of influential observation in linear regression , 2000 .

[10]  Bo-Cheng Wei,et al.  Case-deletion measures for models with incomplete data , 2001 .

[11]  Felix Famoye,et al.  Zero-Inflated Generalized Poisson Regression Model with an Application to Domestic Violence Data , 2021, Journal of Data Science.

[12]  X L Meng,et al.  The EM algorithm and medical studies: a historical linik , 1997, Statistical methods in medical research.

[13]  K. Yau,et al.  Zero‐Inflated Negative Binomial Mixed Regression Modeling of Over‐Dispersed Count Data with Extra Zeros , 2003 .

[14]  Andy H. Lee,et al.  Zero‐inflated Poisson regression with random effects to evaluate an occupational injury prevention programme , 2001, Statistics in medicine.

[15]  Alfred A. Bartolucci,et al.  A rich family of generalized Poisson regression models with applications , 2005, Math. Comput. Simul..

[16]  Diane Lambert,et al.  Zero-inflacted Poisson regression, with an application to defects in manufacturing , 1992 .

[17]  Alan Agresti,et al.  Random effect models for repeated measures of zero-inflated count data , 2005 .

[18]  Chih-Ling Tsai,et al.  Assessing local influence in linear regression models with first-order autoregressive or heteroscedastic error structure , 1992 .

[19]  S. Ferrari,et al.  Beta Regression for Modelling Rates and Proportions , 2004 .

[20]  J. Hinde,et al.  A Score Test for Testing a Zero‐Inflated Poisson Regression Model Against Zero‐Inflated Negative Binomial Alternatives , 2001, Biometrics.

[21]  Dankmar Böhning,et al.  The zero‐inflated Poisson model and the decayed, missing and filled teeth index in dental epidemiology , 1999 .

[22]  Andy H. Lee,et al.  Multi-level zero-inflated Poisson regression modelling of correlated count data with excess zeros , 2006, Statistical methods in medical research.

[23]  D. Pregibon.  Logistic Regression Diagnostics , 1981 .

[24]  R. Cook.  Assessment of Local Influence , 1986 .