Nonparametric estimation in heteroskedastic regression

Michael G. Akritas

Research output: Contribution to journalArticlepeer-review


We consider the problem of making inferences about the parameters in a heteroskedastic regression model using the ranks of weighted observations. The model assumes symmetric error distribution and a parametric model for the error variance. It is shown that there is no loss in asymptotic efficiency due to estimating the unknown weights. This extends the theory of rank estimation in the heteroskedastic linear model.

Original languageEnglish (US)
Pages (from-to)23-31
Number of pages9
JournalStatistics and Probability Letters
Issue number1
StatePublished - Jun 1 1996

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty


Dive into the research topics of 'Nonparametric estimation in heteroskedastic regression'. Together they form a unique fingerprint.

Cite this