Open AccessMean Value Estimation Using Low Size Samples Extracted from Skewed Populations
Author(s) -
João Paulo Martins,
Miguel Felgueiras,
Rui Santos
Publication year - 2022
Publication title -
wseas transactions on mathematics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.211
H-Index - 21
eISSN - 2224-2880
pISSN - 1109-2769
DOI - 10.37394/23206.2022.21.16
Subject(s) - statistics , sample size determination , statistic , confidence interval , statistical inference , inference , mathematics , sample (material) , coverage probability , population , estimation , population mean , econometrics , computer science , artificial intelligence , demography , management , sociology , economics , chemistry , chromatography , estimator
The use of the T-statistic in statistical inference procedures is usually restricted to normal populations or to large samples. However, these conditions may not be fulfilled in some situations: the population can be moderate/highly skewed, or the sample size can be small. In this work, we use the Pearson’s system of distributions, namely, type IV distributions to model T. By some simulation work, it is shown that this approximation leads to confidence intervals whose coverage is close to the nominal coverage even for low sample sizes.