Showing 1 - 10 of 11
Persistent link: https://www.econbiz.de/10010509830
This note gives an easily verified necessary and sufficient condition for one probability forecaster to empirically outperform another one in terms of all strictly proper scoring rules.
Persistent link: https://www.econbiz.de/10010306244
We consider 1927 borrowers from 54 countries who had a credit rating by both Moody's and S&P as of the end of 1998, and their subsequent default history up to the end of 2002. Viewing bond ratings as predicted probabilities of default, we show that it is unlikely that both agencies are well...
Persistent link: https://www.econbiz.de/10010306287
We generalize the refinement ordering for well calibrated probability forecasters to the case were the debtors under consideration are not necessarily identical. This ordering is consistent with many well known skill scores used in practice. We also add an illustration using default predictions...
Persistent link: https://www.econbiz.de/10011451384
The paper explores the relationship between various orderings among probability forecasts that have been suggested in the literature. It is shown that well calibrated forecasters are in general not comparable according to the domination ordering suggested by Vardeman and Meeden (1983), that the...
Persistent link: https://www.econbiz.de/10010316440
We generalize the refinement ordering for well calibrated probability forecasters to the case were the debtors under consideration are not necessarily identical. This ordering is consistent with many well known skill scores used in practice. We also add an illustration using default predictions...
Persistent link: https://www.econbiz.de/10011432495
The paper explores the relationship between various orderings among probability forecasts that have been suggested in the literature. It is shown that well calibrated forecasters are in general not comparable according to the domination ordering suggested by Vardeman and Meeden (1983), that the...
Persistent link: https://www.econbiz.de/10009770515
We consider 1927 borrowers from 54 countries who had a credit rating by both Moody s and S&P as of the end of 1998, and their subsequent default history up to the end of 2002. Viewing bond ratings as predicted probabilities of default, we show that it is unlikely that both agencies are well...
Persistent link: https://www.econbiz.de/10010509832
We consider 1927 borrowers from 54 countries who had a credit rating by both Moodys and S&P as of the end of 1998, and their subsequent default history up to the end of 2002. Viewing bond ratings as predicted probabilities of default, we show that it is unlikely that both agencies are well...
Persistent link: https://www.econbiz.de/10009295178
This note gives an easily verified necessary and sufficient condition for one probability forecaster to empirically outperform another one in terms of all strictly proper scoring rules.
Persistent link: https://www.econbiz.de/10009295185