Interview Question: Weighted Regressions

Sample Question #195 (statistics – regressions)

How does the weighted least squares (WLS) regression method work? What’s its biggest empirical drawback?

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

One Response to Interview Question: Weighted Regressions

  1. Brett says:

    ANSWER
     
    WLS uses a "weight matrix" to derive a consistent least squares estimator; WLS can be seen as a specific example of the GLS method. The weights are meant to "correct" certain problems in the data, such as non-spheric errors. Usually, WLS is carried out by using the inverses of the moment variances as the weights.
     
    The biggest problem can be seen as follows: when the variance of a moment is very small, its inverse will be large. If this happens due to some data measurement error, we inadvertently let this measurement error dominate the WLS estimation. This is a serious defect that every researcher must be aware of.
     

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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