Interview Question: Varying Variance

Sample Question #135 (econometrics)

What is heteroskedasticity? How do you test for the presence of heteroskedasticity?

(Comment: as far as I know, heteroskedasticity, alternatively spelled as heteroscedasticity, is only discussed in econometrics, not in general statistics; I wonder why)

This entry was posted in Sample Qs. Bookmark the permalink.

2 Responses to Interview Question: Varying Variance

  1. Brett says:

    Heteroskedasticity means the error term in a regression model has varying variances instead of a constant variance, violating one of the assumptions behind the OLS model.  (Bonus question: what does this do to the OLS estimator?)[Bonus answer: it does not change the unbiasedness or consistency of the OLS estimator, but gives the wrong estimate for the OLS estimator’s standard deviation, thus leading the researcher to the wrong conclusion about the statistical significance of the OLS estimator.]
    Some of the tests are the White Test, the Breusch-Pagan Test and the Goldfeld-Quandt Test.  (Bonus question: how does each of these tests work? What assumptions about the form of the heteroskedasticity are behind each test procedure?)

  2. Brett says:

    BTW, each of the test, if detecting the presence of heteroskedasticity, also suggests a specific way to "fix" the incorrect stats on the linear estimator.  In general, we can use a weighted least squares (WLS) estimator.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s