Linear Programming in Exploratory Data Analysis

Ronald D. Armstrong, Edward L. Frome and Michael G. Sklar

Journal of Educational and Behavioral Statistics, 1980, vol. 5, issue 4, 293-307

Abstract: It has long been popular to utilize the least Squares estimation procedure for fitting the multiple linear regression model to observed data. In this paper, two useful alternatives to least Squares (L 2 norm) estimation in exploratory data analysis are examined: least absolute value estimation (L 1 norm) and Chebychev (L ∞ norm) estimation. Formulating the L 1 norm and L ∞ norm problems as linear programming problems offers several advantages, including efficient Solution methods using special-purpose Computer codes. An example is provided in which the three procedures are used to fit a line, both with and without an outlier present in the data.

Keywords: Linear programming; Data analysis; Least squares; Least absolute value; Chebychev estimation (search for similar items in EconPapers)
Date: 1980
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.sagepub.com/doi/10.3102/10769986005004293 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sae:jedbes:v:5:y:1980:i:4:p:293-307

DOI: 10.3102/10769986005004293

Access Statistics for this article

More articles in Journal of Educational and Behavioral Statistics
Bibliographic data for series maintained by SAGE Publications ().