Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/195386 
Year of Publication: 
2017
Citation: 
[Journal:] Revista de Métodos Cuantitativos para la Economía y la Empresa [ISSN:] 1886-516X [Volume:] 24 [Publisher:] Universidad Pablo de Olavide [Place:] Sevilla [Year:] 2017 [Pages:] 169-189
Publisher: 
Universidad Pablo de Olavide, Sevilla
Abstract (Translated): 
It is known that, when in the linear regression model there is a high degree of multicollinearity, the results obtained by using the Ordinary Least Squares (OLS) method are unstable. As a solution to this situation, in this paper we present the raised method, the ridge method and the orthogonal variables method as an alternative to the estimate by OLS. It is also shown that regression with orthogonal variables makes sense regardless of the existence of serious multicollinearity because it allows to answer questions which are not accessible when using the original model. These methodologies are applied to a data set of yields of treasury bills.
Subjects: 
regression models
multicollinearity
raised regression
ridge regression
regression with orthogonal variables
JEL: 
C13
C21
Creative Commons License: 
cc-by-sa Logo
Document Type: 
Article

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.