Bayesian structural time series

Bayesian methods

The Bayesian structural time series (BSTS) model is a statistical technique used for feature selection, time series forecasting, nowcasting, inferring causal impact and other applications. The model is designed to work with time series data.

Difference-in-differences models[1] and interrupted time series designs[2] are alternatives to this approach.

The model consists of three main components:

1. Kalman filter. The technique for time series decomposition. In this step, a researcher can add different state variables: trend, seasonality, regression, and others.
2. Spike-and-slab method. In this step, the most important regression predictors are selected.
3. Bayesian model averaging (ensemble learning). Combining the results and prediction calculation.