AVALIAÇÃO DO DESEMPENHO DE ESTIMADORES DOS PARÂMETROS DA DISTRIBUIÇÃO PARETO CORRIGIDOS POR BOOTSTRAP

Francisca Mendonça Souza, Pollyanna Kelly de Oliveira Silva, Marcelo Bourguignon Pereira, Leandro Carlos de Souza

Abstract


This paper evaluates empirically the performance of parameters estimatorsrelated to the Pareto distribution. From samples generated by this distribution,we calculated punctual perform estimations of the parameters usingthe maximum likelihood method, method of moments and their respectiveversions corrected by bootstrap. Many different scenarios were considered,specially the sample size used. We can add that the estimators were evaluatedconsidering their variances, mean squared errors and bias. In general,the estimators corrected by bootstrap showed better results compared tothe uncorrected estimators, i. e., when the estimator is biased.



DOI: https://doi.org/10.5902/2179460X9351

Refbacks

  • There are currently no refbacks.


Copyright (c)



Creative Commons License

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