Forecasting
Overfitting
SPA
MCS
In-Sample Fit and Out-of-Sample Fit: Their Joint Distribution and Its Implications for Model Selection
This paper shows that in-sample overfit translates into out-of-sample underfit - one-to-one. This has important implications for standard model selection criteria, such as AIC and BIC.
This paper shows that in-sample overfit translates into out-of-sample underfit - one-to-one. This has important implications for standard model selection criteria, such as AIC and BIC.
Tests designed to compare multiple forecasting Models
Test for Superior Predictive Ability
Test for Superior Predictive Ability
Qrinkage: Criteria-Based Shrinkage
