Beta
ManageSearchChatDiscover
44839

STATISTICAL ESTIMATION OF EFFECTS OF OBSERVATIONS AND RESIDUAL ANALYSIS ON LINEAR MODEL WITH APPLICATION ON EGYPTIAN INSURANCE MARKET.

Article

Last updated: 04 Jan 2025

OverviewSimilar Items

Subjects

-

Tags

-

Abstract

The Parameters of a linear model are usually estimated by the method of least squares which is sensitive to large values of the additive error term. The advent of modern computers has led to alternative methods to least squares go under the robust regression, these methods are insensitive to deviations from normal distributions and to the effects of outliers in the data. The main purpose of this is to study the problems caused by the presence of outliers. To achieve this aim, there different approaches: M-estimators, Generalized Shrunken M- estimators, Least absolute deviation, S-estimators. Bootstrap Regression depend on re-sampling bootstrap, Robust Bootstrap Regression.   The research has been applied to the Egyptian insurance market marine function. This function consists of total premium (y) , total indemaities (x1), agent commission (x2) , administrative expenses(x3) Summary and conclusions:- The discarding of the outlying and influential observations approach cannot be immediately recommended unless there is strong evidence, which indicates that it resulted from a mistake.    Robust regression analysis provides an alternative to a least squares regression model when fundamental assumptions are unfulfilled by the nature of the data. Other robust estimation methods have been suggested. When fundamental regression analysis assumptions are violated, the researcher may wish to consider what alternatives are available to him. If the data contain influential outliers, then he may wish to employ some form or robust regression that down weights the influence of the troublesome outliers. Will may compare his robust results to his classical results.  

DOI

10.21608/jpd.2008.44839

Keywords

Statistical estimation, residual analysis, linear model, Egyptian insurance market

Authors

First Name

Mohamed

Last Name

Aly

MiddleName

-

Affiliation

Applied Statistics & Insurance Department, Suez Canal University, Egypt.

Email

-

City

Suez Canal University

Orcid

-

First Name

Soltan

Last Name

Abd ElHamid

MiddleName

-

Affiliation

Applied Statistics & Insurance Department, Mansoura University, Manoura, Egypt.

Email

-

City

Mansoura University

Orcid

-

First Name

Khedr

Last Name

Ramadan

MiddleName

-

Affiliation

Productive Efficiency Institute of Zagazig University, Zagazig, Egypt.

Email

-

City

Zagazig

Orcid

-

Volume

13

Article Issue

1

Related Issue

6946

Issue Date

2008-01-01

Receive Date

2007-11-25

Publish Date

2008-01-01

Page Start

229

Page End

250

Print ISSN

1110-2543

Online ISSN

2682-3322

Link

https://jpd.journals.ekb.eg/article_44839.html

Detail API

https://jpd.journals.ekb.eg/service?article_code=44839

Order

15

Type

Original Article

Type Code

867

Publication Type

Journal

Publication Title

Journal of Productivity and Development

Publication Link

https://jpd.journals.ekb.eg/

MainTitle

STATISTICAL ESTIMATION OF EFFECTS OF OBSERVATIONS AND RESIDUAL ANALYSIS ON LINEAR MODEL WITH APPLICATION ON EGYPTIAN INSURANCE MARKET.

Details

Type

Article

Created At

22 Jan 2023