Piecewise Regression
- 1 September 1970
- journal article
- application
- Published by Informa UK Limited in Journal of the American Statistical Association
- Vol. 65 (331), 1109-1124
- https://doi.org/10.1080/01621459.1970.10481147
Abstract
A difficult regression parameter estimation problem is posed when the data sample is hypothesized to have been generated by more than a single regression model. To find the best-fitting number and location of underlying regression systems, the investigator must specify both the statistical criterion and the search-estimation procedure to be used. The approach outlined in this article is essentially a wedding of hierarchical clustering and standard regression theory. As the name suggests, piecewise regression may be described as a method of finding that piecewise continuous function which best describes the data sample. Computational procedures and a fully-worked example, together with possible extensions, are provided.Keywords
This publication has 9 references indexed in Scilit:
- A Dynamic Programming Algorithm for Cluster AnalysisOperations Research, 1969
- Curve Fitting by Segmented Straight LinesJournal of the American Statistical Association, 1969
- Hierarchical clustering schemesPsychometrika, 1967
- Fitting Segmented Curves Whose Join Points Have to Be EstimatedJournal of the American Statistical Association, 1966
- Estimates for the Points of Intersection of Two Polynomial RegressionsJournal of the American Statistical Association, 1964
- Hierarchical Grouping to Optimize an Objective FunctionJournal of the American Statistical Association, 1963
- Tests of Equality Between Sets of Coefficients in Two Linear RegressionsEconometrica, 1960
- Hierarchical Linkage Analysis for the Isolation of TypesEducational and Psychological Measurement, 1960
- The Estimation of the Parameters of a Linear Regression System Obeying Two Separate RegimesJournal of the American Statistical Association, 1958