Ctrip Technology
Sep 12, 2023 · Artificial Intelligence
Using BSTS and CausalImpact for Causal Effect Estimation in Structured Time‑Series Data
The article explains how Bayesian Structured Time Series (BSTS) combined with the CausalImpact library can be used to estimate causal effects for policies or marketing interventions when traditional A/B experiments are infeasible, detailing model theory, Bayesian inference, MCMC estimation, code implementation, and a real‑world holiday‑push case study.
BSTSBayesian ModelingCausalImpact
0 likes · 20 min read