00000ctm a22000004a 4500 UP-99796217608823100 Buklod 20090506165411.0 m o j ta 090506s2006 xx d r |||| u| (iLib)UPD-00085897104 DML eng DMLUC LG 995 2006 S8 S52 Sibbaluca, Vivian Malayao 1981- Generalized method of moments estimation of errors-in-variables in dynamic models by Vivian Malayao Sibbaluca. 2006. [v], 72 leaves ill. "June 2006." Vita. Thesis (M.S. Statistics)--University of the Philippines Diliman. Measurement error is pervasive in survey data. Its existence in the explanatory variables violates basic assumption in regression model which lead to biased and inconsistent Ordinary Least Squares (OLS) estimates. Instrumental variables estimation is typically used to address the problem, which paved way for the development of the Generalized Method of Moments (GMM) estimation. Dynamic model was considered in this study which, like Errors-in-Variables (EIV) models, violates the exogeneity assumption in regression. Measurement errors, also referred to as EIV, are incorporated in the model for which as system GTMM estimator was derived, particularly Blundell-Bond estimator for EIV in dynamic model which is referred to as system estimator in this study. The system estimator was applied to swine production data and was compared to the existing Blundell-Bond estimator which does not assume measurement errors. Results of this study show that for the swine production data, there is little percent difference between the parameters estimates of the two estimators both in the first step and the second step. Its standard errors, however, differ significantly in the first step where these are higher when measurement error is present in the data. Hence, measurement errors affect only the first step standard errors for the said data. Moments method (Statistics) Error analysis (Mathematics) Estimation theory. FI UP Thesis UPD DARCHIVES LG 995 2006 S8 S52 UPD DSTC LG 995 2006 S8 S52 Thesis