## Model Types |
## BATS/TBATSBATS and TBATS models are variations of an exponential smoothing state space model with a Box-Cox statistical transformation using ARMA (autoregressive and moving average) errors. A TBATS model incorporates a trigonometric representation of seasonal components into the model (De Livera et. all 2010). A BATS or TBATS model could also be considered a time series decomposition method that allows multiple complex seasonalities to be incorporated simultaneously. For example, there may be a weekly seasonal component and a monthly seasonal component which both need to be incorporated into the forecast model. BATS and TBATS models can handle this situation and incorporate time series decomposition methodology. Thus, BATS and TBATS models could be considered a hybrid method.
A time series decomposition model consists of decomposing a time series into trend, seasonal, cyclical, and irregular components. Then each component is explicitly estimated and measured statistically. Each estimated component is then recombined in order to estimate a final model and calculate predictions going forward. A BATS model will then incorporate additional autoregressive and moving average components (p, q) to better model any leftover patterns presented in the model residuals. |