COMPSTAT 2023: Start Registration
View Submission - COMPSTAT2023
A0173
Title: Unbiased estimators of the cumulants under bi-additive models Authors:  Sandra Ferreira - University of Beira Interior, Covilha (Portugal) [presenting]
Dario Ferreira - University of Beira Interior (Portugal)
Abstract: The focus is on the mixed-effects linear model, a topic that has been extensively researched in both theory and practice, particularly when the normal distribution is considered. We now consider the case where the random component of the model is assumed to follow a Gamma distribution. We focus on the challenges of using the fourth cumulant, including the need for more accurate and reliable estimates. We then discuss the estimators of the cumulants, which involve linear combinations of first-order cumulants with randomly chosen coefficients. To conclude, we present a simulation study and analyze its results in detail.