After one has created their out-of-sample forecasts, one naturally wants to know how well their models would have historically performed. To facilitate the error analysis and forecast comparison process, OOS contains the forecast_accuracy and forecast_comparison functions.

Forecast accuracy
1. Mean Square Error (MSE)
2. Root Mean Square Error (RMSE)
3. Mean Absolute Error (MAE)
4. Mean Absolute Percentage Error (MAPE)

Notes: All loss functions may estimated simultaneously with the OOS forecast_accuracy function, for all forecasts models present in a forecast_univariate, forecast_multivariate, or forecast_combine created long-form data.frame.

Forecast comparison
1. Forecast Error Ratios
2. Diebold-Mariano Test (for unnested models)
3. Clark and West Test (for nested models)

Notes: Forecast comparisons may be created with the OOS forecast_comparison function, for all forecasts models present in a forecast_univariate, forecast_multivariate, or forecast_combine created long-form data.frame.