Electronic Journal of Statistics
- Electron. J. Statist.
- Volume 9, Number 1 (2015), 1424-1448.
The Signed-rank estimator for nonlinear regression with responses missing at random
This paper is concerned with the study of the signed-rank estimator of the regression coefficients under the assumption that some responses are missing at random in the regression model. Strong consistency and asymptotic normality of the proposed estimator are established under mild conditions. To demonstrate the performance of the signed-rank estimator, a simulation study is conducted under different settings of model error’s distributions, and shows that the proposed estimator is more efficient than the least squares estimator whenever the error distribution is heavy-tailed or contaminated. When the model error follows a normal distribution, the simulation experiment shows that the signed-rank estimator is more efficient than its least squares counterpart whenever a large proportion of the responses are missing.
Electron. J. Statist., Volume 9, Number 1 (2015), 1424-1448.
Received: June 2014
First available in Project Euclid: 29 June 2015
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Bindele, Huybrechts F. The Signed-rank estimator for nonlinear regression with responses missing at random. Electron. J. Statist. 9 (2015), no. 1, 1424--1448. doi:10.1214/15-EJS1042. https://projecteuclid.org/euclid.ejs/1435584427