A summary of some estimators of dynamic panel data models and their applications
Item
-
Title
-
A summary of some estimators of dynamic panel data models and their applications
-
Identifier
-
d_2009_2013:8994e85cc4ba:11445
-
identifier
-
11834
-
Creator
-
Ma, Zhen,
-
Contributor
-
Michael Grossman
-
Date
-
2012
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Economics
-
Abstract
-
This thesis consists of two chapters. Chapter one summarizes three estimators of dynamic panel data models: Generalized Method of Moments (GMM) with fixed effects, Wooldridge Conditional Maximum Likelihood (CML) with random effects and a Maximum Simulated Likelihood (MSL) random effects dynamic probit. Chapter two presents their applications and empirical findings. I examine the impact of the large price increases in cigarettes after the Master Settlement Agreement (MSA) on drinking behavior using data from the Panel Study of Income Dynamics (PSID). Alcohol consumption, drinking participation and heavy drinking participation (three or more drinks per day) are considered for the full sample, as well as for sub-samples stratified by age group and gender. Estimation results are relatively stable across estimators. I find that the cross-price effects of cigarettes on alcohol consumption are insignificant showing that averaging on all consumption levels, the number of drinks consumed per day is not affected by the increases in cigarette prices; and that the cross-price effects of cigarettes on drinking participation are mostly positive and significant, indicating drinking is an economic substitute for smoking; also, cigarette prices do not affect heavy drinking prevalence.
-
Type
-
dissertation
-
Source
-
2009_2013.csv
-
degree
-
Ph.D.
-
Program
-
Economics
