Abstract
The variance-covariance matrix for the seemingly unrelated regression estimator is expressed as an ordinary-least-squares variance-covariance matrix. This permits new insights into the efficiency of seemingly unrelated regression, especially the role of correlations among variables.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 137-139 |
| Number of pages | 3 |
| Journal | American Statistician |
| Volume | 42 |
| Issue number | 2 |
| DOIs | |
| State | Published - May 1988 |
Keywords
- Multicollinearity
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty
- General Mathematics