A Multivariable Regression Algorithm
Abstract
A BASIC algorithm has been developed that is capable of fitting a user-defined regression equation to a set of data. This best-fit-curve algorithm is unique in that it allows multiple variables and multiple forms (exponential, trigonometric, logarithmic, etc.) to be present in a single regression equation. The least-squares regression performed determines the constants for each of the regression equation terms to provide a best-fit curve. Other programs within the algorithm set allow for data entry, editing and print-out, and plotting of the raw data and their best-fit regression curve.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1983
- Accession Number
- ADA136630
Entities
People
- George L. Blaisdell
- Todd Carpenter
Organizations
- Cold Regions Research and Engineering Laboratory