PERFORMANCES OF MODIFIED POWER DIVERGENCE ESTIMATORS IN THE NORMAL MODELS : SIMULATION AND COMPARATIVE STUDY

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F12%3A00199565" target="_blank" >RIV/68407700:21340/12:00199565 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

PERFORMANCES OF MODIFIED POWER DIVERGENCE ESTIMATORS IN THE NORMAL MODELS : SIMULATION AND COMPARATIVE STUDY

Popis výsledku v původním jazyce

The main interest of our research was to examine these modifications in practical use as to the consistency, robustness and efficiency of the estimators. We focus on the well known family of power divergences parametrized by ff alpha in R in the normal distribution model. First we study the robustness by theoretical means deriving the influence functions and asymptotic variances, then we present the results of extensive computer simulation study for several randomly selected contaminated and uncontaminated data sets, and we study the behavior of estimators for different sample sizes and different -divergence parameters. According to our results, the superdivergence estimators are not very interesting for detailed scrutiny and we focus on the subdivergence estimators which show a considerable robustness. Finally we compare these estimators with other, mostly robust to some extent, estimators such as median, median absolute deviation, minimum Kolmogorov distance estimators, minimum power

Název v anglickém jazyce

PERFORMANCES OF MODIFIED POWER DIVERGENCE ESTIMATORS IN THE NORMAL MODELS : SIMULATION AND COMPARATIVE STUDY

Popis výsledku anglicky

The main interest of our research was to examine these modifications in practical use as to the consistency, robustness and efficiency of the estimators. We focus on the well known family of power divergences parametrized by ff alpha in R in the normal distribution model. First we study the robustness by theoretical means deriving the influence functions and asymptotic variances, then we present the results of extensive computer simulation study for several randomly selected contaminated and uncontaminated data sets, and we study the behavior of estimators for different sample sizes and different -divergence parameters. According to our results, the superdivergence estimators are not very interesting for detailed scrutiny and we focus on the subdivergence estimators which show a considerable robustness. Finally we compare these estimators with other, mostly robust to some extent, estimators such as median, median absolute deviation, minimum Kolmogorov distance estimators, minimum power

Druh

O - Ostatní výsledky

CEP obor

BA - Obecná matematika

OECD FORD obor

—

Projekt

<a href="/cs/project/LG12020" target="_blank" >LG12020: Využití pokročilé statistické analýzy a nestatistických separačních metod pro detekování fyzikálních procesů v datech snímaných urychlovači elementárních částic.</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2012

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů