Exponential smoothing models never forget past data as long as the smoothing constant is strictly between 0 and 1. In contrast, moving average methods completely forget all data older than k periods in the past.

Question

Quizplus · Accepted Answer

Exponential smoothing models use a weighted average of past data, where the weights decrease exponentially as the observations get older. This means that past data still influences the forecast, but to a lesser extent as it becomes farther in the past. Moving average methods, on the other hand, give equal weight to the last k periods and completely ignore any data older than k periods.

Exponential Smoothing Models Never Forget Past Data as Long as the Smoothing