Estimating a bivariate linear relationship

David Leonard

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

Solutions of the bivariate, linear errors-in-variables estimation problem with unspecified errors are expected to be invariant under interchange and scaling of the coordinates. The appealing model of normally distributed true values and errors is unidentified without additional information. I propose a prior density that incorporates the fact that the slope and variance parameters together determine the covariance matrix of the unobserved true values but is otherwise diffuse. The marginal posterior density of the slope is invariant to interchange and scaling of the coordinates and depends on the data only through the sample correlation coefficient and ratio of standard deviations. It covers the interval between the two ordinary least squares estimates but diminishes rapidly outside of it. I introduce the R package leiv for computing the posterior density, and I apply it to examples in astronomy and method comparison.

Original languageEnglish (US)
Pages (from-to)727-754
Number of pages28
JournalBayesian Analysis
Volume6
Issue number4
DOIs
StatePublished - 2011

Keywords

  • Errors-in-variables
  • Identifiability
  • Measurement error
  • Straight line fitting

ASJC Scopus subject areas

  • Statistics and Probability
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Estimating a bivariate linear relationship'. Together they form a unique fingerprint.

Cite this