Métodos bootstrap: uma aplicação em dados olímpicos via python

Jalmar Carrasco; Lizandra Castilho Fabio; Luciano Santana

Authors

Jalmar Carrasco Universidade Federal da Bahia
Lizandra Castilho Fabio Federal University of Bahia
Luciano Santana Federal University of Bahia

Keywords:

Bootstrap resampling methods, Olympic games, Generalized linear model

Abstract

Bootstrap methods have occupied an important place in the world of statistics due to their simplicity and computational power. Over the past decade, extensions as Bayesian bootstrap, double, bootknife have been proposed. In practice, the maximum likelihood method is widely used to find estimators with good proprieties; however for some situations this estimator has a little bias, particularly when the sample size is small. To avoid this problem, bootstrap techniques can be used to correct the bias. A set of data from the Olympic Games is used to exemplify the studied methodology.

Downloads

Download data is not yet available.

References

AKAIKE, H. A new look at the statistical model identification. IEEE Transactions on Automatic Control. v.19, p. 716-723, 1974.

ANDERSON, T. W. e DARLING, D. A. Asymptotic theory of certain "goodness-of-fit" criteria based on stochastic processes. Annals of Mathematical Statistics. v.23, p. 193–212, 1952.

CIRILLO, M. A. et al. Avaliação de métodos de estimação intervalar para funções lineares binomiais via Bootstrap finito. Ciência e Agrotecnologia. v, 33, p, 1741 -1746, 2009.

CHEN, P.Y. e POPOVICH, P.M. Correlation: parametric and nonparametric measures. Thousand Oaks: Sage Publication, 2002.

CLÍMACO, G. N. Otimização da extração de compostos bioativos da beterraba por metodologia de superfície de resposta e método de Bootstrap. 2019. Dissertação (mestrado em Engenharia de Alimentos )-Universidade Estadual de Maringá.

DAVIDSON, R. e MACKINNON, J. G. Improving the reliability of bootstrap tests with the fast double bootstrap. Computational Statistics Data Analysis. v.51, p. 3259-3281, 2007.

DOGAN, C. D. Applying bootstrap resampling to compute confidence intervals for

various statistics with R. Eurasian Journal of Educational Research. v.68, p. 1-17, 2017.

EDGARD MATSUKI. Brasil sobe de 37º para 35º no ranking histórico das Olimpíadas. Agência Brasil, c2016. Disponível em: https://agenciabrasil,ebc,com,br/rio-2016/noticia/2016-08/brasil-sobe-de-37o-para-35o-no-quadro-de-medalhas-com-19-conquistadas-no. Acesso em: 08 de jul. de 2021

EFRON, B. Bootstrap methods: another look at the jackknife. The annals of Statistics. v.7, p. 1–26, 1979.

EFRON, B. (1983). Estimating the error rate of a prediction rule: improvement on cross-validation, Journal of the American Statistical Association. v.78, p. 316–331, 1983.

EFRON, B. e TIBSHIRANI, R. J. An introduction to the bootstrap. Chapman and Hall: New York, 2014.

GILES, H. e MENTCH, L. Bootstrap bias corrections for ensemble methods. Statistics and Computing. V. 28, p. 77-86, 2018.

HESTERBERG, T. C. Smoothed bootstrap and jackboot sampling. MathSoft. Inc. Seattle, 1999.

KAGGLE. Your Machine Learning and Data Science Community, Disponível em: https://kaggle,com. Acesso em: 08 de jul. de 2021

LIMA, F. P. Inferência bootstrap em modelos de regressão beta. 2017. Tese (doutorado Estatística) - Universidade Federal de Pernambuco.

MCCULLAGH, P. e NELDER, J. A. Generalized Linear Models. Chapman and Hall: London, 1989.

MOREIRA, G. R. F. Aplicação de método estocástico no cálculo das provisões de sinistros. 2020. Trabalho de conclusão de curso de graduação - Universidade Federal de São Paulo.

NELDER, J. A. e WEDDERBURN, R. W. M. Generalized linear models. Journal of the Royal Statistical Society, v.135, p. 370-384, 1972.

PAULA, G. A. Modelos de regressão com apoio computacional. Instituto de Matemática e Estatística, Universidade de São Paulo, 1993.

R CORE TEAM. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna: Austria, 2021.

ROSSUM, G. V. e DRAKE, F. L. Python Tutorial. Python Software Foundation, 2012.

RUBIN, D. J. The bayesian bootstrap. The annals of Statistics. v.9, p. 130–134, 1981.

TANG, G. e LI, J. Regression analysis-based chinese olympic games competitive sports strength evaluation model research. The Open Cybernetics Systemics Journal. v.9, p. 2729–2735, 2015.

THE WORLD BANK. Is a unique global partnership fighting poverty worldwide through sustainable solutions. Disponível em: https://data,worldbank,org/. Acessado em: 10 de jul, de 2021.

WIT, E. et al. ’all models are wrong,,,’: an introduction to model uncertainty. Statistica Neerlandica, v.66, p. 217–236, 2012.