Comparison of the Goodness-of-Fit Tests for Truncated Distributions

Autor:	Agnieszka Lach, Łukasz Smaga
Rok vydání:	2019
Předmět:	010104 statistics & probability Goodness of fit 0502 economics and business 05 social sciences Statistics General Medicine 050207 economics 0101 mathematics 01 natural sciences Mathematics
Zdroj:	Przegląd Statystyczny. 65:296-313
ISSN:	2657-9545 0033-2372
DOI:	10.5604/01.3001.0014.0541
Popis:	The aim of this paper is to investigate the finite sample behavior of seven goodness-of-fit tests for left truncated distributions of Chernobai et al. (2015) in terms of size and power. Simulation experiments are based on artificial data generated from the distributions that were used in the past or are used nowadays to describe the tails of asset returns. The study was conducted for different tail thickness and for changing truncation point. Simulation results indicate that the testing procedures do not work equally well under finite samples, and some of them require quite large number of observations to perform satisfactorily.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::1710bcd84d33e95511d4c16dc73ae0d3 https://doi.org/10.5604/01.3001.0014.0541 Zobrazit plný text záznamu