What does RMSE measure in forecast accuracy?

What does RMSE measure in forecast accuracy?

• A. The autocorrelation of residuals.
• B. The proportion of explained variance.
• C. The bias of forecasts.
• D. The average magnitude of forecast errors.

Answer: (D)

RMSE measures the typical size of forecast errors, in the same units as the variable being forecast. It’s calculated as the square root of the average of the squared differences between forecasts and actual values. Because errors are squared before averaging, larger mistakes weigh more, so RMSE highlights big deviations and lower values indicate better accuracy. This focus on the average magnitude of errors makes it the go-to single-number summary of forecast accuracy. It’s different from bias (the average error direction), and from explained variance or residual autocorrelation, which look at different aspects of forecast performance.

RMSE measures the typical size of forecast errors, in the same units as the variable being forecast. It’s calculated as the square root of the average of the squared differences between forecasts and actual values. Because errors are squared before averaging, larger mistakes weigh more, so RMSE highlights big deviations and lower values indicate better accuracy. This focus on the average magnitude of errors makes it the go-to single-number summary of forecast accuracy. It’s different from bias (the average error direction), and from explained variance or residual autocorrelation, which look at different aspects of forecast performance.