Comparação de Métodos de Estimação em um Modelo Linear Simples com Erro nas Variáveis

Tiago Almeida de Oliveira, Augusto Ramalho de Moraes, Marcelo Angelo Cirillo

Abstract


Although the researcher is very careful errors experiment may bepresent, therefore, is appropriate to evaluate the reliability of data, mainlywhen different estimation methods are used. Considering this problem,the objective of this work is compare the performance of two methods ofestimation known as “Plug-in” and “Attenuation bias”. These methodsare used in the regression model with errors-in-variables techniques. TheMonte-Carlo simulations are used for generated a linear regression modelerrors-in-variables, different sample sizes, residuals distributions and lackof fit were considered. It was possible to conclude that, “Plug-in “ and “Attenuation bias” methods were accurate with R2=90%, however, thesemethods have better accuracy for symmetrical distributions.



DOI: https://doi.org/10.5902/2179460X9483

Copyright (c)



Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

 

DEAR AUTHORS,

PLEASE, CHECK CAREFULLY BEFORE YOUR SUBMISSION:

1. IF ALL AUTHORS "METADATA" (ORCID, LINK TO LATTES, SHORT BIOGRAPHY, AFFILIATION) WERE ADDED,

2. THE CORRECT IDIOM YOUR SECTION,

3 IF THE HIGHLIGHTS WERE ADDED,

4. IF THE GRAPHIC ABSTRACTS WAS ADDED,

5. IF THE REVIEWERS INDICATION WAS DONE,

6. IF THE REFERENCES FORMAT ARE CORRECT(ABNT)

7. IF THE RESOLUTION YOUR FIGURES (600 DPI) ARE SUITABLE

8.  IF THE STATEMENT BY THE ETHICS COMMITTEE (IF IT INVOLVES HUMANS) WAS ADDED;

9. IF THE DECLARATION OF ORIGINALITY WAS ADDED.

10. IF THE TEXT IS ORIGINAL. IF THE IDEA HAS ALREADY BEEN REGISTERED IN SUMMARY FORM, OR PUBLISHED IN CONGRESS ANNUALS, PLEASE INFORM THE EDITOR.