Doctoral Dissertations
Abstract
"In this dissertation, a statistical test for endogeneity of a binary independent variable in a simultaneous system of two equations is constructed. The value taken by the binary variable is assumed to depend on the unobserved value of one of the dependent variables, as is the case in some common econometric formulations, such as those suggested by Heckman. The endogeneity is assumed to arise through the correlation of the error terms of the two equations. The asymptotic distribution of the test statistic is derived and the sma11 sample properties of the test are studied using Monte Carlo methods. The simulation results show that the significance level of the test remains close to the nominal value and the test provides good power in detecting deviations from the null hypothesis of exogeneity, even when sample size is as small as fifty. Wu, Hausman and Dhrymes have established several tests for detecting endogeneity of a continuous independent variable, but a test for detecting the endogeneity of a binary independent variable that has been derived on a sound theoretical basis is currently not available. Thus, the proposed test, which is based on a generalization of the test developed by Drymes for the continuous case, fills a gap in the literature related to endogeneity tests"--Abstract, page iii.
Advisor(s)
Samaranayake, V. A.
Committee Member(s)
Bryant, Richard
Drain, David
Gelles, Gregory M.
Gadbury, Gary L.
Department(s)
Mathematics and Statistics
Degree Name
Ph. D. in Mathematics
Publisher
University of Missouri--Rolla
Publication Date
Fall 2006
Pagination
viii, 124 pages
Note about bibliography
Includes bibliographical references (pages 122-123).
Rights
© 2006 Xu Cao, All rights reserved.
Document Type
Dissertation - Restricted Access
File Type
text
Language
English
Subject Headings
Econometric modelsEquations, SimultaneousMonte Carlo method
Thesis Number
T 9056
Print OCLC #
168623279
Recommended Citation
Cao, Xu, "A likelihood ratio test of endogeneity for binary independent variables" (2006). Doctoral Dissertations. 1709.
https://scholarsmine.mst.edu/doctoral_dissertations/1709
Share My Dissertation If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.
