StatREC is the product of teamwork between Nikolaos Mittas, Ioannis Mamalikidis and Lefteris Angelis.
The first version (v1.0) of StatREC was presented on Promise 2012 entitled: “StatREC: A Graphical User Interface Tool for Visual Hypothesis Testing of Cost Prediction Models”, published by ACM; view by clicking here.
Motivated by its success, whilst the name remains as is, we’ve enriched the software with enough new functions to make it all but a new programme.
Version 2.x offers a variety of analysis methods, as well as the option to compute Local and Global errors and the Mean Model too. Full Changelog Here
Visit the STA.IN.S research group webpage by clicking here
StatREC is based on the conceptual idea of Regression Error Characteristic (REC) Curves.
It supports the practical application of all visualisation, statistical and hypothesis procedures needed for comparing Cost Prediction Models.
One has access, amongst others, to the Permutation Distribution, the REC Curve, the REC Replicates, the AOC REC Curve, the Mean and Median, the observed kolmogorov-Smirnov statistic, the t-test value, the wilcoxon value, and more.