Electronic Journal of Statistics

Weighted least squares estimation with missing responses: An empirical likelihood approach

Anton Schick

Full-text: Open access


A heteroscedastic linear regression model is considered where responses are allowed to be missing at random. An estimator is constructed that matches the performance of the weighted least squares estimator without the knowledge of the conditional variance function. This is usually done by constructing an estimator of the variance function. Our estimator is a maximum empirical likelihood estimator based on an increasing number of estimated constraints and avoids estimating the variance function.

Article information

Electron. J. Statist., Volume 7 (2013), 932-945.

First available in Project Euclid: 3 April 2013

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62F12: Asymptotic properties of estimators 62G05: Estimation
Secondary: 62J05: Linear regression

Heteroscedastic linear regression missing at random maximum empirical likelihood estimation estimated constraints increasing number of constraints efficiency


Schick, Anton. Weighted least squares estimation with missing responses: An empirical likelihood approach. Electron. J. Statist. 7 (2013), 932--945. doi:10.1214/13-EJS793. https://projecteuclid.org/euclid.ejs/1364994253

Export citation


  • [1] Carroll, R. J. (1982). Adapting for heteroscedasticity in linear models., Ann. Statist. 10 1224–1233.
  • [2] Chamberlain, G. (1987). Asymptotic efficiency in estimation with conditional moment restrictions., J. Econometrics 34 305–334.
  • [3] Chen, S.X., Peng, L. and Qin, Y.-L. (2009). Effects of data dimension on empirical likelihood., Biometrika 96 711–722.
  • [4] Hjort, N.L., McKeague, I.W. and Van Keilegom, I. (2009). Extending the scope of empirical likelihood., Ann. Statist. 37 1079–1111.
  • [5] Koul, H., Müller, U.U. and Schick, A. (2012). The transfer principle: a tool for complete case analysis., Ann. Statist. 60 3031–3047.
  • [6] Little, R.J.A. and Rubin, D.B. (2002)., Statistical Analysis with Missing Data. Second edition. Wiley Series in Probability and Statistics, Wiley, Hoboken.
  • [7] Müller, H. G. and Stadtmüller, U. (1987). Estimation of heteroscedasticity in regression analysis., Ann. Statist. 15 610–625.
  • [8] Müller, U.U. and Van Keilegom, I. (2012). Efficient parameter estimation in regression with missing responses., Electron. J. Statist. 6 1200–1219.
  • [9] Owen, A. B. (1988). Empirical likelihood ratio confidence intervals for a single functional., Biometrika 75 237–249.
  • [10] Owen, A. B. (1990). Empirical likelihood ratio confidence regions., Ann. Statist. 18 90–120.
  • [11] Owen, A.B. (2001)., Empirical Likelihood. Chapman & Hall/CRC.
  • [12] Peng, H. and Schick, A. (2012a). Empirical likelihood approach to goodness of fit testing. To appear in, Bernoulli.
  • [13] Peng, H. and Schick, A. (2012b). Maximum empirical likelihood estimation and related topics. Preprint available at, http://math.binghamton.edu/anton/preprint.html.
  • [14] Qin, J. and Lawless, J. (1994). Empirical likelihood and general estimating equations., Ann. Statist. 22 300–325.
  • [15] Robinson, P. M. (1987). Asymptotically efficient estimation in the presence of heteroskedasticity of unknown form., Econometrica 55 875–891.
  • [16] Schick, A. (1987). A note on the construction of asymptotically linear estimators., J. Statist. Plann. Inference 16, 89–105. Correction (1989) 22 269–270.
  • [17] Tsiatis, A.A. (2006)., Semiparametric Theory and Missing Data. Springer Series in Statistics. Springer, New York.