What is a common rule for selecting a forecasting model using error measures?

What is a common rule for selecting a forecasting model using error measures?

• A. Always choose the model with the most parameters to capture complexity.
• B. Compare forecast accuracy using metrics like MAPE, RMSE; select model with lowest error while avoiding overfitting.
• C. Use the model with the highest R-squared regardless of overfitting.
• D. Choose the model with the smallest training error only.

Answer: (B)