10 Popular Machine Learning Time Series Model
3 min readFeb 12, 2023
- Autoregressive Integrated Moving Average (ARIMA) — ARIMA is a statistical model that is widely used for time series forecasting. It models the time series data as a combination of an autoregression (AR) component, which models the relationship between the current time step and past time steps, and a moving average (MA) component, which models the residual errors from a prior model. ARIMA is especially useful for modeling time series data with a clear underlying trend and seasonality.
- Exponential Smoothing (ETS) — ETS is a family of models for time series forecasting that use weighted averages of historical data to make predictions. The weights decrease exponentially as the data becomes older, allowing for more recent data to have a greater impact on the forecast. ETS models can be further classified as simple exponential smoothing, Holt’s linear exponential smoothing, and Holt-Winters exponential smoothing, depending on the complexity of the model and the number of variables being considered.
- Seasonal Decomposition of Time Series (STL) — STL is a method for decomposing a time series into its seasonal, trend, and residual components. By separating the time series into these components, STL allows for more accurate forecasting by modeling each component separately and then recombining them. STL is particularly useful for time series data with a clear underlying seasonality.
- Long Short-Term Memory (LSTM) — LSTM is a type of recurrent neural network (RNN) that is well suited for time…
