What would you like to understand?

In exponential smoothing, the first period's forecast is typically initialized with the first period's actual value ($F_1 = A_1$). Because the forecast and the actual value are identical, the error is zero and uninformative.

Conditions: The initial forecast is set to the first actual value ($F_1 = A_1$).; You are calculating forecast errors to evaluate model accuracy.

In exponential smoothing, the first period's forecast is typically initialized with the first period's actual value ($F_1 = A_1$). Because the forecast and the actual value are identical, the error is zero and uninformative.

Conditions: The initial forecast is set to the first actual value ($F_1 = A_1$).; You are calculating forecast errors to evaluate model accuracy.