Model Fitting

'STEPIT' is a Fortran II subroutine for finding local minima of almost any real function (called CHISQ) of several variables (called X(I)). The X(I) might, for example, be parameters appearing nonlinearly in an expression which is to be fitted to data by minimizing chi-square. We use this routine for most of our model fitting. Included in the SHAR file are the source code for STEPIT, sample programs, utility subroutines and shells, and some documents.

STEPIT shar distribution file
STEPIT documentation
STEPIT tutorial
Additional modeling examples from Massaro, D. W., Cohen, M. M, & Gesi, A. T. (1993). Long-term training, transfer, and retention in learning to lipread. Perception & Psychophysics, 53, 549-562.
PSL Data Warehouse including various STEPIT models

Model Fitting

This page has been accessed 8014 times since Tue Jan 6 16:12:21 PST 1998.