|
|
| (One intermediate revision by one other user not shown) |
| Line 1: |
Line 1: |
| In [[statistics]], the '''White test''' is a [[statistical test]] that establishes whether the [[errors and residuals in statistics|residual]] [[variance]] of a variable in a [[regression model]] is constant: that is for [[homoscedasticity]].
| | Jayson Berryhill is how I'm known as and my wife doesn't like it at all. As a lady what she really likes is fashion and she's been free tarot readings - [http://www.seekavideo.com/playlist/2199/video/ http://www.seekavideo.com/playlist/2199/video] - doing it for quite a whilst. Credit authorising is how he tends to make cash. North Carolina is exactly where we've been residing for many years and will by no means move.<br><br>Also visit my web page free [http://alles-herunterladen.de/excellent-advice-for-picking-the-ideal-hobby/ good psychic] reading [[http://gcjcteam.org/index.php?mid=etc_video&document_srl=696611&sort_index=regdate&order_type=desc her response]] |
| | |
| This test, and an estimator for [[heteroscedasticity-consistent standard errors]], were proposed by [[Halbert White]] in 1980.<ref>{{cite journal
| |
| |last=White |first=H. |authorlink=Halbert White
| |
| |year=1980
| |
| |title=A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity | |
| |journal=[[Econometrica]]
| |
| |volume=48 |issue=4 |pages=817–838
| |
| |mr=575027 | jstor = 1912934
| |
| }}</ref> These methods have become extremely widely used, making this paper one of the most cited articles in economics.<ref>{{cite journal
| |
| |title=What Has Mattered to Economics since 1970
| |
| |journal=Journal of Economic Perspectives
| |
| |volume=20 |number=4 |year=2006 |pages=189–202
| |
| |last1=Kim |first1=E.H.
| |
| |last2=Morse |first2=A.
| |
| |last3=Zingales |first3=L.
| |
| |doi=10.1257/jep.20.4.189
| |
| |url=http://faculty.chicagogsb.edu/finance/papers/What%20Has%20Mattered%20to%20Economics%20Since%201970.pdf
| |
| }}</ref>
| |
| | |
| ==Testing constant variance==
| |
| To test for constant variance one undertakes an auxiliary regression analysis: this regresses the squared residuals from the original regression model onto a set of [[regressor]]s that contain the original [[regressor]]s, the cross-products of the regressors and the squared regressors. One then inspects the <math>R^{2}</math>. The [[Lagrange multiplier test|Lagrange multiplier (LM) test]] statistic is the product of the ''R''<sup>2</sup> value and sample size:
| |
| :<math>\ LM = n \cdot R^2 .</math>
| |
| This follows a [[chi-squared distribution]], with degrees of freedom equal to the number of estimated parameters (in the auxiliary regression).
| |
| | |
| An alternative to the White test is the [[Breusch–Pagan test]].
| |
| | |
| If homoscedasticity is rejected one can use [[heteroscedasticity-consistent standard errors]].
| |
| | |
| ==See also== | |
| *[[Heteroscedasticity]]
| |
| | |
| ==References==
| |
| {{reflist}}
| |
| | |
| [[Category:Statistical tests]]
| |
| [[Category:Regression diagnostics]]
| |
| | |
| {{statistics-stub}}
| |
Jayson Berryhill is how I'm known as and my wife doesn't like it at all. As a lady what she really likes is fashion and she's been free tarot readings - http://www.seekavideo.com/playlist/2199/video - doing it for quite a whilst. Credit authorising is how he tends to make cash. North Carolina is exactly where we've been residing for many years and will by no means move.
Also visit my web page free good psychic reading [her response]