Smoothing Methods
"Smoothing" relies on averaging values over multiple periods in order to reduce the noise and uncover the patterns. Smoothing forecasting methods are data-driven as they estimate time series components (Error, Trend and Seasonality) directly from the data, without a predetermined structure. This is convenient in series where the components change over time. They can be automated, although the user must select the smoothing constants to define how fast the methods adapt to the new data.
Moving Average
Exponential Smoothing
ETS models compute forecasts as weighted averages of past observations, with the weights decaying exponentially for older observations.