Detecting outliers and influential observations with heteroscedasticity-corrected models
Heteroscedasticity-correction masks signals from standardized residuals, so analysts should examine the residuals to identify outliers and should use likelihood dispersion to identify influential observations. These points are demonstrated with a model that examines the effect of exchange rate volatility on intra-industry trade.
Year of publication: |
2005
|
---|---|
Authors: | Martin, David ; Kumar, Vikram |
Published in: |
Applied Economics Letters. - Taylor & Francis Journals, ISSN 1350-4851. - Vol. 12.2005, 12, p. 745-748
|
Publisher: |
Taylor & Francis Journals |
Saved in:
freely available
Saved in favorites
Similar items by person
-
Detecting outliers and influential observations with heteroscedasticity-corrected models
Martin, David, (2005)
-
Inter-Temporal Purchasing Power Parity
Boucher Breuer, Janice,
-
Thinking Outside the Pipeline : Venturing into Distributed Off-Grid Water Markets
Kumar, Vikram, (2010)
- More ...