Accession Number : ADA113387

Title :   Restricted L1 Estimators and Their Covariances,

Descriptive Note : Technical rept.,

Corporate Author : TEXAS A AND M UNIV COLLEGE STATION INST OF STATISTICS

Personal Author(s) : Book,D ; Booker,J ; Hartley,H O ; Sielken,R L , Jr

PDF Url : ADA113387

Report Date : Jun 1980

Pagination or Media Count : 57

Abstract : The parameters in a linear regression model can be estimated by minimizing the sum of the absolute residuals (L1 estimation) instead of the more classical approach of minimizing the sum of squared residuals (least squares estimation). In addition to other nice properties L1 estimators are less sensitive to outliers than least squares estimators. This paper describes a linear programming algorithm and computer program for obtaining L1 estimators and estimates of their covariances when the regression parameters are restricted to satisfy specified linear constraints. These estimated covariances are the new feature in this work and are an extremely important ingredient in hypothesis tests and confidence interval construction.

Descriptors :   *Linear regression analysis, *Linear programming, *Covariance, Models, Algorithms, Computer programs, Estimates, Residuals, Least squares method, Parameters, Monte Carlo method, Input output models, Sampling

Subject Categories : Statistics and Probability
      Computer Programming and Software

Distribution Statement : APPROVED FOR PUBLIC RELEASE