google/CausalImpact

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1854 stars · 264 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

CausalImpact is an R package for estimating the causal effect of a designed intervention on a time series using Bayesian structural time-series models. It estimates how a response metric would have evolved without the intervention, useful when randomized experiments are not available.

Use cases

• measure the impact of an ad campaign on daily clicks
• estimate causal effect of a policy change on a time series
• run counterfactual analysis when no A/B test is possible
• analyze how a product launch affected website traffic
• estimate intervention effects using control time series
• quantify incremental impact of a marketing intervention

When to choose

• you need to estimate causal effects on time series without a randomized experiment
• you have unaffected control time series to model the counterfactual
• you work in R and want a well-established, maintained package

When to avoid

• you cannot identify control series unaffected by the intervention
• the relationship between treated and control series is unstable over the post-period
• you need a Python-native implementation (use TFP CausalImpact instead)

Facets

library · maturity stable

machine-learning data-science analytics data-science analytics time-series python causal-inference bayesian-structural-time-series time-series counterfactual-analysis intervention-analysis r-package statistics r

Member repositories

For agents

markdown · JSON · MCP: product_card(name="google/CausalImpact")