Monte Carlo Investigation of Robust Methods

R. Dutter and I. Ganster
Additional contact information
R. Dutter: Technical University Graz, Institute of Statistics
I. Ganster: Technical University Graz, Institute of Statistics

A chapter in Probability and Statistical Inference, 1982, pp 59-72 from Springer

Abstract: Abstract The need and importance of statistical methods which should behave well not only at the supposed parametric model but also in a certain neighbourhood, are obvious. Such “robust” methods are often difficult to investigate analytically. Well-developed simulation techniques with high-speed computers enable us to examine quantitatively up to a certain degree the behaviour of such procedures. In this paper we describe some possibilities for the investigation of robust statistical methods by Monte Carlo techniques. As an illustration the examination of location estimators when different symmetric distributions (long- and short-tailed) are underlying, is reported. 62 estimators are considered where most of them are designed to guard against outliers in a certain way. Distributions are mainly taken from the exponential power family. Some analysis of the simulated mean square errors is also given.

Date: 1982
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-94-009-7840-9_8

Ordering information: This item can be ordered from
http://www.springer.com/9789400978409

DOI: 10.1007/978-94-009-7840-9_8

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().