Regress: A Biased Regression Package.

Abstract

REGRESS is a linear multiple regression package consisting of programs capable of providing the computations necessary for Ordinary Least Squares Regression Analysis (OLS), Latent Root Regression Analysis (LRRA), Principle Component Regression Analysis (PCR), Ridge Regression Analysis (RRA), and a Baranchik (James-Stein) Shruken Estimator (BSE). The user may call any or all of these analyses. The purpose of Regress is to bring OLS and four methods of biased regression together in one user-oriented package so the user can more easily analyze data which may be subject to multicollinearities. (Author)

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 1976
Accession Number
ADA033441

Entities

People

  • Robert F. Pierce

Organizations

  • Southern Methodist University

Tags

DTIC Thesaurus Topics

  • Computational Complexity
  • Computations
  • Data Science
  • Estimators
  • Information Science
  • Mathematical Analysis
  • Mathematics
  • Regression Analysis
  • Statistical Algorithms
  • Statistical Analysis

Fields of Study

  • Mathematics

Readers

  • Computer Science.
  • Regression Analysis.