1
$\begingroup$

Consider an OLS model with $n$ observations and $p$ explanatory variables (including an intercept term) $$y=X\beta + \epsilon$$ We may use a White test to (approximately) check for the presence of heteroscedasticity on an $\alpha$-significance level, by conducting an auxiliary regression: $$e = Y\gamma + \eta$$ where $e=y-X\widehat\beta$ ($\widehat\beta$ is the OLS estimator of $\beta$) and $Y$ is the matrix of all the auxiliary regression variables $1, X_i, X_i\odot X_j$. The Lagrange multiplier statistic $nR^2$ of this auxiliary regression is then compared against $\chi_\alpha^2(p)$. If $nR^2>\chi_\alpha^2(p)$, the null hypothesis that $\epsilon$ is homoscedastic is rejected.

Now suppose we believe $\epsilon$ is heteroscedastic with $\Bbb V(\epsilon\mid X) = \Lambda$ where $\Lambda$ is a positive diagonal matrix. Then we may use a WLS model to reestimate the original OLS by specifying the weight $W:=\Lambda^{-1}$, and premultiply the original model by $W^{1/2}$ to obtain $$W^{1/2}y=W^{1/2}X\beta+W^{1/2}\epsilon$$ and the WLS estimator of $\beta$ which is an OLS estimator of $\beta$ under the transformed model should be $$\widehat\beta_{W}=(X^TWX)^{-1}X^TWy.$$ Question: how do we use a White test to measure how accurately $W$ has captured the real heteroscedasticity structure?

A naive thought would be just using the WLS residual $e_W:=y-X\widehat\beta_W$ and the original $Y$ as the inputs for the White test. However, it is evident that $$\Bbb E(\widehat\beta_W)=(X^TWX)^{-1}X^TW\Bbb E(y)=(X^TWX)^{-1}X^TWX\beta=\beta=\Bbb E(\widehat\beta)$$ so $\widehat\beta_W$ usually shouldn't deviate too much from $\widehat\beta$, and thus $e_W$ won't be much different from $e$. Hence such a way of "White test" can tell little about how well $\Bbb V(\epsilon\mid X)$ is addressed by $W$.

Another more plausible thought is consider the transformed model instead of the original OLS model, i.e. use the weighted residual $W^{1/2}e_W$ for the auxiliary regression. But when it comes to choosing the explanatory variables we'll have a problem: should we choose the original $Y$ (spanned by zero, first and second degree terms of $X$)? It seems weird matching $W^{1/2}e_W$. So a more natural idea would be matching $e_W$ with $W^{1/2}X$. But wait, there's no longer an intercept term in $W^{1/2}X$, and there seems to be no way to naturally construct auxiliary variables from $W^{1/2}X$ without introducing a new dof..

So how is this done in reality?

$\endgroup$
1
$\begingroup$

Given you know $W$, then let $Y_{new} = W^{1/2}Y$ and $X_{new} = W^{1/2}X$. Then forget about $Y$ and $X$. Go back to use $Y_{new}$ and $X_{new}$ to perform OLS, to use a White test to (approximately) check for the presence of heteroscedasticity.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks but my problem lie exactly in how to use X_new for the White test? The original intercept term is lost when premultiplied by $W^{1/2}$, and if we "forcefully" introduce a new intercept term, well, we effectively introduce a new dof in $X_{new}$.... $\endgroup$ – Vim Nov 2 '18 at 17:44
  • $\begingroup$ How do you lost intercept? In addition, white test does not require intercept. Suppose you have a intercept and 4 slopes in the Y-X model. In your new model $Y_{new}-X_{new}$, you still have 5 regression coefficients, right? $\endgroup$ – user158565 Nov 2 '18 at 17:51
  • $\begingroup$ $W^{1/2}X$ no longer has an intercept column (the initial intercept column in $X$ is turned into a non constant column namely varies with rows). $\endgroup$ – Vim Nov 3 '18 at 6:08
  • $\begingroup$ It is OK, just use $X_{new}$ . The number of columns in $X_{new}$ and $X$ are the same. The first column of $X$ is 1, and the first column of $X_{new}$ is not only 1. But it does not matter. $\endgroup$ – user158565 Nov 3 '18 at 6:13
  • $\begingroup$ actually it does matter. Theoretically, White test requires an intercept to be used. Also if I only pass $X_{new}$ to the white test function of a statistical package (say Python statsmodels.api) then an Error occurs saying the model dof doesn't match with the rank of $Y_{new}$. $\endgroup$ – Vim Nov 3 '18 at 6:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.