Keywords

polynomial curve fitting, data, uncertainty, least squares, regression

Abstract

This article reviews the theory and some good practice for fitting polynomials to data. I show by theory and example why fitting using a basis of orthogonal polynomials rather than monomials is desirable. I also show how to scale the independent variable for a more stable fit. I also demonstrate how to compute the uncertainty in the fit parameters. Finally, I discuss regression analysis: how to determine whether adding an additional term to the fit is justified.

Document Type

Peer-Reviewed Article

Publication Date

2018-09-01

Language

English

College

Physical and Mathematical Sciences

Department

Physics and Astronomy

University Standing at Time of Publication

Full Professor

Share

COinS