Suojin Wang, Editor-in-Chief, Journal of Nonparametric Statistics

The May 2010 issue features a discussion paper, “Identification and Estimation of Nonlinear Models Using Two Samples with Nonclassical Measurement Errors,” by Raymond Carroll, Xiaohong Chen, and Yingyao Hu. As the title indicates, this paper considers identification and estimation of a general nonlinear errors-in-variables model using two samples.

Special Rate for Members
Journal of Nonparametric Statistics is now a publication of the American Statistical Association. ASA members are entitled to a personal online subscription for $15 per volume. Click here to order.

This statistical problem is especially appealing in current research in econometrics, biology, and medicine. The authors show that the general latent nonlinear model is nonparametrically identified using the two samples when both could have nonclassical errors, without either instrumental variables or independence between the two samples. In addition, the authors propose sieve quasi maximum likelihood estimation for the parameter of interest and investigate its properties when the two samples are independent and the nonlinear regression model is parameterized.

Five experts—Aurore Delaigle, Peter Hall, Han Hong, Marie-Luce Taupin, and Young Truong—contribute interesting discussions from both theoretical and practical viewpoints. They offer additional insight and ideas for further research and development. These discussions are followed by a rejoinder.

The discussion paper, comments, and rejoinder will be freely available online until December 31.

In addition to the discussion and rejoinder papers, seven other papers appear in this issue. These papers cover a range of research interests in nonparametric statistics, from functional estimation to variable selection to goodness-of-fit testing. Click here to view the full table of contents.