jakobrunge/tigramite

Health v2 · maintenance only

Adoption not part of the score

1710 stars · 319 forks observed · 2026-08-28

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

Tigramite is a Python package for causal inference on time series data, providing causal discovery methods such as PCMCI, PCMCIplus, LPCMCI, and RPCMCI together with a range of conditional independence tests. It also supports causal effect estimation, mediation analysis, robust forecasting from learned graphs, and plotting of results.

Use cases

• estimate causal graphs from time series data
• discover lagged and contemporaneous causal links in autocorrelated datasets
• estimate direct, total, and mediated causal effects from a known causal graph
• run conditional independence tests on continuous or discrete time series
• analyze regime-dependent causal relationships
• perform causal time series analysis in climate or geoscience research

When to choose

• you need causal discovery specifically designed for time series with lags
• you want a choice of linear and non-parametric conditional independence tests
• you need to handle hidden confounders or multiple datasets in causal discovery
• you want to estimate causal effects or mediation from time series graphs

When to avoid

• you need causal discovery for purely cross-sectional (non-time-series) data
• you want a general-purpose machine learning forecasting library without causal structure
• you need a GUI-based causal analysis tool rather than a Python API

Facets

library · maturity active

machine-learning data-science math data-science machine-learning python cross-platform causal-inference causal-discovery time-series conditional-independence-tests pcmci causal-effects algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="jakobrunge/tigramite")