Exact inference for the linear model with groupwise heteroscedastic spherical disturbances
Exact inference on a single coefficient in a linear regression model, as introduced by Bekker (1997), is elaborated for the case of normally distributed heteroscedastic disturbances. Instead of approximate inference based on feasible generalized least squares, exact confidence sets are formulated based on partial rotational invariance of the distribution of the vector of disturbances. The approach is applied to the random-effects and fixed-effects models for panel data.
Year of publication: |
2000
|
---|---|
Authors: | Bekker, Paul ; Leertouwer, Erik |
Institutions: | Faculteit Economie en Bedrijfskunde, Rijksuniversiteit Groningen |
Saved in:
freely available
Saved in favorites
Similar items by person
-
Who creates poltical business cycles? : (should central banks be blamed?)
Leertouwer, Erik, (1999)
-
International and domestic constraints on political business cycles in OECD economies: a comment
Leertouwer, Erik, (1999)
-
Roo, Gert de, (2004)
- More ...