Publication detail

Bootstrap Estimation of Standard Error of Coefficient

Authors: Linda Bohdan | Kubanová Jana

Year: 2005

Type of publication: ostatní - přednáška nebo poster

Name of source: ISBIS4 Conference

Publisher name: The international society for Business and Industrial Statistics

Place:

Page from-to:

Titles:

Language	Name	Abstract	Keywords
cze	Bootstrapový odhad standardní chyby koeficientu	Článek řeší problém užití lineárního modelu v případech, kdy nejsou splněny základní podmínky jeho fungování. Ukazuje na řešení pomocí resamplingových metod.	regresní model, odhad parametrů
eng	Bootstrap Estimation of Standard Error of Coefficient	The linear regression models have a great importance in technical and economic practice. In case when random errors of the model have normal probability distribution and their variance does not change, it is possible to use the method of the least squares to estimate the coefficients of regression model, which is at this case exact method. If random errors in the model have other probability distribution than normal one or if their variance changes depending on changing values of declarative variable, the problem how to find exact method for determination of the properties of estimation of regression coefficients arises. The asymptotical methods based on central limit theorem are usually used, or another approximations. Such methods require usually complicated mathematical apparatus and their conclusions can be usually less reliable with regard to small number of data. Relatively simple method, that can help us to solve problems like this, is called bootstrap. This paper deals with estimation of standard error of estimation of the regression coefficient of the linear regression model using above mentioned method.	regression model, parameters estimate

Search

Login for students

Login for employees

Publication detail

Map

Services

Popular Links