Avaliação de métodos para comparação de modelos de regressão por simulação de dados

Sérgio Ricardo Silva Magalhães; Ruben Delly Veiga; Thelma Sáfadi; Marcelo Ângelo Cirillo; Márcio Magini

doi:10.5433/1679-0375.2004v25n2p117

Evaluation of methods for comparing regression models by data simulation

Authors

Sérgio Ricardo Silva Magalhães Universidade Vale do Rio Verde de Três Corações
Ruben Delly Veiga Universidade Federal de Lavras
Thelma Sáfadi Universidade Federal de Lavras
Marcelo Ângelo Cirillo Universidade Federal de Lavras
Márcio Magini Universidade do Vale do Paraíba

DOI:

https://doi.org/10.5433/1679-0375.2004v25n2p117

Keywords:

Models Identity, Dummy variables (binary), Variance analysis.

Abstract

The present study was intended to evaluate the statistic methods of Models Identity, Dummy Variables (binary) and Variance Analysis, used for comparing regression models, by means of computer data simulation. Four linear regression cases and five cases of quadratic polynomial regression were considered. By utilizing the resources of the modulus of the Interactive Matrix Language (IML) of the Statistical Analysis System (SAS), appropriate routines were developed for the implementation of the methodology. A data simulation made up of 10.000 experiments was carried out, considering the different sample sizes (10, 50 and 100 observations) for each one of the nine cases reported. The results of all the situations simulated through the three methods were similar, presenting a low Type I Error and Type II Error percentage. The Dummy Variable Method proved to be the most efficient for the three sizes of samples, for it presented the lowest percentages of Type I Error and Type II Error. Key words: Models Identity. Dummy Variables (binary). Variance Analysis.

Downloads

Download data is not yet available.

Author Biographies

Sérgio Ricardo Silva Magalhães, Universidade Vale do Rio Verde de Três Corações

Professor Assistente I – UNINCOR - Universidade Vale do Rio Verde de Três Corações – Campus Betim e Belo Horizonte.

Ruben Delly Veiga, Universidade Federal de Lavras

Professor Titular do Departamento de Ciências Exatas da UFLA.

Thelma Sáfadi, Universidade Federal de Lavras

Professora Adjunta do Departamento de Ciências Exatas da UFLA.

Marcelo Ângelo Cirillo, Universidade Federal de Lavras

Doutorando em Estatística – UFLA.

Márcio Magini, Universidade do Vale do Paraíba

Professor Títular do Instituto de Pesquisa e Desenvolvimento - UNIVAP.

Downloads

PDF (Português (Brasil))

Published

2004-12-15

How to Cite

Magalhães, S. R. S., Veiga, R. D., Sáfadi, T., Cirillo, M. Ângelo, & Magini, M. (2004). Evaluation of methods for comparing regression models by data simulation. Semina: Ciências Exatas E Tecnológicas, 25(2), 117–122. https://doi.org/10.5433/1679-0375.2004v25n2p117

Download Citation

Issue

Vol. 25 No. 2 (2004)

Section

Original Article

License

The Copyright Declaration for articles published in this journal is the author’s right. Since manuscripts are published in an open access Journal, they are free to use, with their own attributions, in educational and non-commercial applications. The Journal has the right to make, in the original document, changes regarding linguistic norms, orthography, and grammar, with the purpose of ensuring the standard norms of the language and the credibility of the Journal. It will, however, respect the writing style of the authors. When necessary, conceptual changes, corrections, or suggestions will be forwarded to the authors. In such cases, the manuscript shall be subjected to a new evaluation after revision. Responsibility for the opinions expressed in the manuscripts lies entirely with the authors.

This journal is licensed with a license Creative Commons Attribution-NonCommercial 4.0 International.