AN ASYMPTOTIC TEST FOR THE DETECTION OF HETEROSKEDASTICITY

An asymptotic test for heteroskedasticity has been developed. The test does not rely on any assumption about heteroskedasticity, and introduces two alternative statistics based on the same idea. Power of these two alternative test statistics has been measured by Monte Carlo simulations. For large samples they performed fairly well, whereas for sample sizes ≤ 100, their power was influenced by the structure of the heteroskedasticity.

MoreLess

Year of publication:	2008
Authors:	Yuce, Mehmet
Published in:	Istanbul University Econometrics and Statistics e-Journal. - İktisat Fakültesi. - Vol. 8.2008, 1, p. 33-44
Publisher:	İktisat Fakültesi
Subject:	Heteroskedasticity \| large sample test \| regression analysis \| violations from the assumptions of classical linear regression model \| residual analysis \| asymptotic properties \| Monte Carlo simulations \| the power of the test

freely available

Full text |

More access options

Check Google Scholar

In German libraries (KVK)

I need help

More details

Extent:	application/pdf
Type of publication:	Article
Classification:	C00 - Mathematical and Quantitative Methods. General ; C10 - Econometric and Statistical Methods: General. General ; C19 - Econometric and Statistical Methods: General. Other
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005012093