Regression Error Characteristic Surfaces

DMCA

Regression Error Characteristic Surfaces (2005)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Luís Torgo

Venue:	In Proc. of the 11th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'05), ACM
Citations:	7 - 1 self

BibTeX

@INPROCEEDINGS{Torgo05regressionerror,
    author = {Luís Torgo},
    title = {Regression Error Characteristic Surfaces},
    booktitle = {In Proc. of the 11th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'05), ACM},
    year = {2005},
    publisher = {Press}
}

OpenURL

Abstract

This paper presents a generalization of Regression Error Characteristic (REC) curves. REC curves describe the cumulative distribution function of the prediction error of models and can be seen as a generalization of ROC curves to regression problems. REC curves provide useful information for analyzing the performance of models, particularly when compared to error statistics like for instance the Mean Squared Error. In this paper we present Regression Error Characteristic (REC) surfaces that introduce a further degree of detail by plotting the cumulative distribution function of the errors across the distribution of the target variable, i.e. the joint cumulative distribution function of the errors and the target variable. This provides a more detailed analysis of the performance of models when compared to REC curves. This extra detail is particularly relevant in applications with non-uniform error costs, where it is important to study the performance of models for specific ranges of the target variable. In this paper we present the notion of REC surfaces, describe how to use them to compare the performance of models, and illustrate their use with an important practical class of applications: the prediction of rare extreme values.

Keyphrases

regression error characteristic surface cumulative distribution function rare extreme value important practical class roc curve mean squared error extra detail useful information non-uniform error cost specific range regression error characteristic present regression error characteristic rec curve rec surface joint cumulative distribution function detailed analysis prediction error

DMCA