What is the term for the average of squared forecast errors?

The term that describes the average of squared forecast errors is known as the Mean Squared Error (MSE). This metric is widely used in quantitative analysis to assess the quality of a forecasting model. MSE is calculated by taking the differences between the forecasted and actual values, squaring each of those differences to eliminate any negative values, and then averaging those squared differences.

Using squared errors emphasizes larger discrepancies, thereby giving more weight to significant forecast errors compared to smaller ones. This makes MSE particularly useful when you need a precise measure that takes the magnitude of forecast errors into account. It helps analysts understand how well their model performs and is commonly utilized in regression analysis and model validation processes.

In contrast, the other terms are associated with different calculations and definitions within the context of error measurement and do not specifically refer to the average of squared forecast errors. For instance, the Root Mean Square Error (RMSE) takes the square root of MSE and provides an error measure on the same scale as the original data. Mean average error typically involves averaging the absolute values of forecast errors, and variance of forecast errors refers to how much the errors vary, rather than their average squared value.