The Estimation Of Incomplete Econometric Models With Autogressive Errors Using Subsystem Estimators

Stationary disturbances and asymptotic theory; Specfiml (spectral full information maximum likelihood) estimation; The specfilm estimation with inadequate sample size; The estimation of the multiple regression model with stationary erros and lagged endogenous variables; The specfilm method as applied to models with lagged endogenous variables; The asymptotic variance matrix of the structural estimators when the erros follow an AR process.

This book analyzes how the choice of a particular disclosure limitation method, namely additive and multiplicative measurement error, affects the quality of the data and limits its usefulness for empirical research. Generally, a disclosure limitation method can be regarded as a data filter that transforms the true data generating process. This book focuses explicitly on the consequences of additive and multiplicative measurement error for the properties of nonlinear econometric estimators. It investigates the extent to which appropriate econometric techniques can yield consistent and unbiased estimates of the true data generating process in the case of disclosure limitation. Sandra Nolte received her PhD in Economics at the University of Konstanz, Germany in 2008 and is a postdoctoral researcher at the Financial Econometric Research Centre at the Warwick Business School, UK since 2009. Her research areas include microeconometrics and financial econometrics.