Testing for cointegration tests to see if the residuals from a regression between the two variables is a stationary process. So wouldn't a test on those residuals for heteroskedasticity be the same thing? To detect if the variance of the errors is stable and the error terms are not a random walk? Wouldn't a White test or Breusch-Pagan test for heteroskedasticity be the same as an Augmented Dickey-Fuller test? Both would detect if the errors are getting bigger or have a constant variance, right? An increasing variance would imply a random walk and unit root, correct?

  • 2
    $\begingroup$ Intuitively you are correct. But when you try to get proofs for the statistic under null and alternative then you see the difference. Note also that the null hypotheses are a bit different for tests. For ADF, the null hypothesis is random walk, for BP the null is homoscedasticity. $\endgroup$
    – mpiktas
    Mar 13, 2012 at 7:37
  • 1
    $\begingroup$ A bit more on intuition: formally cointegration implies testing of the residuals on unit roots, that indeed could be viewed as the separate case of heteroscedasticity testing with the equivalent null hypothesis that $\sigma^2_{u_t} = t\cdot c$ with $c$ some constant. All mentioned tests has the null $\sigma^2_{u_t} = c$ with the alternative that has to be chosen and require some other assumptions of the regression model to be true (stochastic regressors uncorrelated with errors, etc.) while cointegration relies on common trends. $\endgroup$ Mar 13, 2012 at 12:06
  • $\begingroup$ I believe ADF is a stationarity test - it tests both mean and variance (whether they are stationary or not). BP only tests for constant variance (homo/hetero skedasticity). $\endgroup$ Nov 20, 2023 at 19:16

1 Answer 1


There isn't much of a resemblance among the stated tests of heteroskedasticity and the cointegration tests, except that you are regressing functions of residuals on other functions of the regressors and the residuals -- but that isn't really saying a lot, as vast number of tests fit that description.

Tests of heteroskedasticity

Consider the linear regression model $$ Y_i = \boldsymbol{X}_i'\boldsymbol{\beta} + \varepsilon_i $$

Based on this regression model there are several regression-based tests of heteroskedasticity -- equivalent test statistics that are not regression-based do exist, but those obviate the comparisons that we are after.


The Breusch-Pagan test of heteroskedasticity has the following steps:

  • Estimate the regression model above using OLS, and get the residuals $\widehat{\varepsilon}_i$, and the standard error of regression, $\widehat{\sigma}^2 =\tfrac{\sum_{i=1}^n\widehat{\varepsilon}_i^2}{n}$.

  • Then, estimate the following auxiliary regression by OLS -- a regression of the standardized residuals on the cross-products of the included regressors.

$$ \boxed{\frac{\widehat{\varepsilon}_i^2}{\widehat{\sigma}^2} = \text{vech}(\boldsymbol{X}_i\otimes\boldsymbol{X}_i')'\boldsymbol{\gamma}+\nu_i} $$

  • The test statistic here is $\tfrac{1}{2}ESS$, which is distributed $\chi^2_{K+\tfrac{K(K+1)}{2}}$, where there are $K$ regressors in the model.


  • The White test is based on a regression that looks very similar to the one employed by BP

$$ \boxed{\widehat{\varepsilon}_i^2 = \text{vech}(\boldsymbol{X}_i\otimes\boldsymbol{X}_i')'\boldsymbol{\gamma}+\nu_i} $$

  • The test statistic here is $nR^2$ which is again distributed $\chi^2_{K+\tfrac{K(K+1)}{2}}$.

Aside: Equivalence of a modified version of BP and White

You would not be mistaken in thinking that there exists a version of the BP test that is exactly equivalent to the White test (which is robust to departures of the residuals from normality). This is discussed in Waldman (1983).

Tests of cointegration

Now consider the Engle-Granger two-step residual-based tests of cointegration.

  • Here, the model is

$$ Y_{1t} = \beta_0 + \boldsymbol{Y}_{2t}'\boldsymbol{\beta} +\varepsilon_{1t} $$ Again, we fit the regresion model using OLS, and get the estimated residuals, $\widehat{\varepsilon}_{1t}$.

  • We now conduct an ADF unit root test on these residuals, that is, we fit the regression $$ \boxed{\Delta \widehat{\varepsilon}_{1t} = \beta_0 + \gamma \widehat{\varepsilon}_{1t-1} + \sum_{j=1}^p\gamma_j \Delta \widehat{\varepsilon}_{1t-j} +\nu_t} $$ and conduct a t-test of the regression coefficient $\gamma=0$ using the Engle-Yoo critical values.


The heteroskedasticity tests regress squares of fitted residuals on regressors, and cointegration tests regress differences of fitted residuals on lags and lags of differences of those residuals (compare the three boxed regressions).

Every model has certain features, each of which can be exploited to form tests of that model. For unit root models, the ADF tests use the specific feature in a specific model -- $\rho=1$ in an autoregressive model -- to test for unit roots. There are other tests, for example, the variance ratio tests that exploit the increasing variance aspect of unit roots. They are all, as you can imagine, related.

  • $\begingroup$ You're really something ... $\endgroup$
    – user14281
    Oct 31, 2012 at 10:28
  • $\begingroup$ This is probably the most concise and practical description I've seen for these. Nice. $\endgroup$ Nov 8, 2014 at 2:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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