Ronald D. Armstrong, E. L. Frome
Abstract
A comparison is made between linear programming and weighted least-squares approaches to absolute deviation curve fitting. The computational results show that linear programming substantially reduces solution times while requiring only a relatively small amount of additional storage.
OpenAlex reports 40 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A comparison is made between linear programming and weighted least-squares approaches to absolute deviation curve fitting. The computational results show that linear programming substantially reduces solution times while requiring only a relatively small amount of additional storage.
Key concepts: Least absolute deviations, Absolute deviation, Mathematics, Algorithm, Least-squares function approximation, Curve fitting, Linear programming, Applied mathematics