The TBATS model is a time series model for series exhibiting multiple complex seasonalities.

The TBATS model was introduced by De Livera, Hyndman & Snyder (2011, JASA). "TBATS" is an acronym denoting its salient features:

The TBATS model is a generalization of the BATS model, which is similar except for lacking the trigonometric regressors.

The TBATS model can be fitted using the tbats() command in the forecast package for R.