Goodness-of-fit patterns in a computer cross-validation procedure comparing a linear and a threshold model

Authors

Charles E. Collyer, University of Rhode IslandFollow

Document Type

Article

Date of Original Version

1986

Abstract

Cross-validation is the process of comparing a model’s predictions to data that were not used in the estimation of model parameters. Cross-validation may have some value in identifying source models, especially in cases where the corresponding fitted models require the estimation of different numbers of parameters. Some of the information available from cross-validation is illustrated using a linear and a threshold model, and goodness-of-fit patterns are contrasted with those of conventional model-fitting.

Citation/Publisher Attribution

Collyer, C. E. Behavior Research Methods, Instruments, & Computers (1986) 18: 618. https://doi.org/10.3758/BF03201437

Available at: https://doi.org/10.3758/BF03201437