Omitted Variable Tests And Dynamic Specification: An Application To Demand Homogeneity by Björn SchmolckOmitted Variable Tests And Dynamic Specification: An Application To Demand Homogeneity by Björn Schmolck

Omitted Variable Tests And Dynamic Specification: An Application To Demand Homogeneity

byBjörn Schmolck

Paperback | July 13, 2000

Pricing and Purchase Info

$113.86 online 
$155.95 list price save 26%
Earn 569 plum® points

Prices and offers may vary in store

Quantity:

In stock online

Ships free on orders over $25

Not available in stores

about

This book deals with the omitted variable test for a multivariate time-series regression model. The empirical motivation is the homogeneity test for a consumer demand system. The consequences of using a dynamically misspecified omitted variable test are shown in detail. The analysis starts with the univariate t-test and is then extended to the multivariate regression system. The small sample performance of the dynamically correctly specified omitted variable test is analysed by simulation. Two classes of tests are considered: versions of the likelihood ratio test and the robust Wald test which is based on a heteroskedasticity and autocorrelation consistent variance-covariance estimator (HAC).
Title:Omitted Variable Tests And Dynamic Specification: An Application To Demand HomogeneityFormat:PaperbackPublished:July 13, 2000Publisher:Springer Berlin HeidelbergLanguage:English

The following ISBNs are associated with this title:

ISBN - 10:354067358X

ISBN - 13:9783540673583

Reviews

Table of Contents

Introduction.- The t-statistic under dynamic misspecification.- The model.- Properties of the estimators.- The distribution of the quasi t-statistic.- Invariance results.- Monte Carlo experimentation.- Consumer theory and the Rotterdam model.- Commodity space and budget set.- Preferences, direct utility function and Marshallian demand.- Cost function and Hicksian demand.- The Rotterdam model.- The Rotterdam model in matrix form.- Robust estimation.- Quasi-maximum likelihood estimation.- Estimation of the covariance matrix of the quasi-maximum likelihood estimator.- Testing for homogeneity.- Anderson`s U test if the errors are time-independent (LRU).- Functional equivalence between the LRU and Laitinen`s statistic.- Likelihood ratio test if the errors are VAR(p).- The asymptotic distribution of the LRU statistic under dynamic misspecification.- The robust Wald test.- Summary.- Monte Carlo experimentation.- Data.- The data-generating process.- Experiments.- Simulation results.- Conclusions.