FitPMM

Author: Cengiz Zopluoglu, Jeffrey R. Haring, & Nidhi Kohli
Contact: https://cengiz.me/fitpmm.html
Description: An R Routine to Fit Finite Mixture of Piecewise Mixed-Effect Models With Unknown Random Knots. In the R routine, a piecewise regression model with an unknown knot is first fitted to each individual's data using the segmented package in R (Muggeo, 2008). Then, multiple start value sets are generated using the procedure outlined by Hipp and Bauer (2006) for the initial stage estimations to overcome the problem of possible local solutions in estimating growth mixture models. After the initial stage estimations for a number of start value sets are completed, the best sets are reestimated until a convergence criterion is met.
Analyses: Item & Test Analysis
System Requirements: Windows, Mac OS, & Linux
Measurement Model: IRT/Rasch
License Type: Open-Source
Documentation: