Nonparametric and Semiparametric Methods in Econometrics and Statistics Barnett

Nonparametric and Semiparametric Methods in Econometrics and StatisticsProceedings of the Fifth International Symposium in Economic Theory and Econometrics\nAuthor(s): William A. Barnett, James Powell, George E. Tauchen\nFormat: Hardback\nPublisher: Cambridge University Press, United Kingdom\nImprint: Cambridge University Press\nISBN-13: 9780521370905, 978-0521370905\nSynopsis\nThis collection of papers delivered at the Fifth International Symposium in Economic Theory and Econometrics in 1988 is devoted to the estimation and testing of models that impose relatively weak restrictions on the stochastic behaviour of data. Particularly in highly non-linear models, empirical results are very sensitive to the choice of the parametric form of the distribution of the observable variables, and often nonparametric and semiparametric models are a preferable alternative. Methods and applications that do not require string parametric assumptions for their validity, that are based on kernels and o.