Hands-On Machine Learning for Algorithmic Trading

Moving averages and exponential smoothing

Early forecasting models included moving-average models with exponential weights called exponential smoothing models. We will encounter moving averages again as key building blocks for linear time series.

Forecasts that rely on exponential smoothing methods use weighted averages of past observations, where the weights decay exponentially as the observations get older. Hence, a more recent observation receives a higher associated weight. These methods are popular for time series that do not have very complicated or abrupt patterns.

Exponential smoothing is a popular technique based on weighted averages of past observations, with the weights decaying exponentially as the observations get older. In other ...