<b>Principal components in the discrimination of outliers: A study in simulation sample data corrected by Pearson's and Yates´s chi-square distance

Autores

  • Manoel Vitor de Souza Veloso Universidade Federal de Alfenas
  • Marcelo Angelo Cirillo Universidade Federal de Lavras

DOI:

https://doi.org/10.4025/actascitechnol.v38i2.26046

Palavras-chave:

contaminated samples, Monte Carlo, significance test, p-value

Resumo

Current study employs Monte Carlo simulation in the building of a significance test to indicate the principal components that best discriminate against outliers. Different sample sizes were generated by multivariate normal distribution with different numbers of variables and correlation structures. Corrections by chi-square distance of Pearson´s and Yates's were provided for each sample size. Pearson´s correlation test showed the best performance. By increasing the number of variables, significance probabilities in favor of hypothesis H0 were reduced. So that the proposed method could be illustrated, a multivariate time series was applied with regard to sales volume rates in the state of Minas Gerais, obtained in different market segments.

 

Downloads

Não há dados estatísticos.

Biografia do Autor

Marcelo Angelo Cirillo, Universidade Federal de Lavras

Prof. Associado I- Departamento de Ciências Exatas - Universidade Federal de Lavras.

Downloads

Publicado

2016-04-01

Como Citar

Veloso, M. V. de S., & Cirillo, M. A. (2016). <b>Principal components in the discrimination of outliers: A study in simulation sample data corrected by Pearson’s and Yates´s chi-square distance. Acta Scientiarum. Technology, 38(2), 193–200. https://doi.org/10.4025/actascitechnol.v38i2.26046

Edição

Seção

Estatí­stica

 

0.8
2019CiteScore
 
 
36th percentile
Powered by  Scopus

 

 

0.8
2019CiteScore
 
 
36th percentile
Powered by  Scopus

Artigos mais lidos pelo mesmo(s) autor(es)