Conditional moment estimation of nonlinear equation systems
This Ph.D. thesis on conditional moment estimation of nonlinear equation systems summarizes the most recent developments on the Generalized Method of Moments Estimator (GMM). In the first part of the book the author reviews several topics of estimation theory and the second part deals with an application of the GMM approach by estimating a model of cooperative R \& D. With respect to the first part of the book the author first introduces the conditional moment approach to GMM estimation and the asymptotic properties of these estimators. Computational aspects are considered in chapter four. These three chapter are quite general and present a good introduction into the theory of GMM estimation. The next chapters deal with more special topics in the context of GMM estimation. Among these are asymptotic efficiency bounds and a discussion of optimal weights as well as optimal instruments. Special emphasis is given to the problem of overidentifying restrictions. The author discusses the asymptotic efficiency gains due to these overidentifying restrictions and moments of compounded distributions. The following two chapters are dedicated to GMM estimation with optimal weights as well as to GMM estimation with optimal instruments. With respect to the first issue, the author looks at iterative estimators and at some shortcomings of IV estimation. In the case of GMM estimation with optimal instruments the chapter presents three alternative estimators, namely a parametric two-step estimation, the \(k\)-nearest neighbor estimator and the kernel estimation. The last chapter of the first part presents the results of some Monte Carlo studies comparing GMM versus ML estimation and versus empirical likelihood estimation. To summarize, the books contains the most recent developments on GMM estimation, discusses several optimality issues and presents applications of the estimation techniques. The book has a clear-cut structure, is sufficiently non-technical for those who are interested only in applications of the method, and it offers most recent estimation techniques for all those researchers dealing with cross section and/or panel data.
- Parameter estimation in systems of nonlinear equations
- On the Approximation of Moments for Nonlinear Systems
- Condition numbers for the nonlinear matrix equation and their statistical estimation
- Statistical Condition Estimation for Linear Systems
- Nonparametric inference based on conditional moment inequalities
- Conditional least squares estimation in nonstationary nonlinear stochastic regression models
- Conditional moments and linear regression for stable random variables
- Estimation of conditional densities and sensitivity measures in nonlinear dynamical systems
This page was built for publication: Conditional moment estimation of nonlinear equation systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1587502)
