What is the difference between the actual time series value and the forecast called?

The difference between the actual time series value and the forecast is known as forecast error. This term is critical in the field of quantitative analysis and predictive modeling, as it quantifies how accurate a forecast is by measuring discrepancies between predicted values and actual outcomes.

Forecast error can be calculated for individual data points or aggregated over a period to provide insights into the performance of a forecasting method. A low forecast error indicates that the model is accurately predicting future values, while a high forecast error suggests that the model may need adjustments or improvements. Understanding forecast error is essential for businesses and analysts who rely on predictions for decision-making, as it helps in evaluating the reliability of the forecasting models in use.

Other terms, such as forecast accuracy, forecast bias, and forecast variance, relate to different aspects of forecasting but do not define the direct difference between actual and predicted values in the same manner. Forecast accuracy encompasses the overall correctness of forecasts, while forecast bias indicates a systematic error in predictions. Forecast variance focuses on the variability of forecast errors rather than their direct calculation.