Causality for time series

Resources

• Granger causality
  • The Granger causality test is a statistical hypothesis test for determining whether one time series is useful in forecasting another, first proposed in 1969
  • Granger causality is a fundamental technique for causal inference in time series data, commonly used in the social and biological sciences
• PCMCI:
  • https://jakobrunge.github.io/tigramite/#tigramite-pcmci-pcmci
  • PCMCI causal discovery for time series datasets. It is a 2-step causal discovery method for large-scale time series datasets. The first step is a condition-selection followed by the MCI conditional independence test.

Bayesian Causal Inference

• Causal analysis with PyMC: Answering “What If?” with the new do operator
• Using Causal ML Instead of A/B Testing
• Benjamin Vincent - What-if- Causal reasoning meets Bayesian Inference | PyData Global 2022
• Causal Inference with CausalPy

Code

• #CODE https://github.com/WillianFuks/tfcausalimpact
  • https://github.com/WillianFuks/tfcausalimpact/blob/master/notebooks/getting_started.ipynb
  • Inferring the effect of an event using CausalImpact by Kay Brodersen
  • https://www.the-odd-dataguy.com/2020/09/09/timeseries-forecasting-and-causal-analysis-in-r-with-facebook-prophet-and-google-causalimpact/
  • https://medium.com/ssense-tech/unveiling-the-benefits-of-causal-inference-in-measuring-advertisement-impact-b1fdcf354d74
• #CODE Tigramite - Tigramite is a time series analysis python module for causal discovery
  • https://jakobrunge.github.io/tigramite/
  • PCMCI algorithm. Causal discovery

References

• #PAPER Causal Generative Neural Networks (2018)
• #PAPER Causal inference using deep neural networks (2020)
• #PAPER #REVIEW Causal Inference for Time series Analysis: Problems, Methods and Evaluation (2021)
• #PAPER High-recall causal discovery for autocorrelated time series with latent confounders (2021)