Causal Inference for Time series Analysis: Problems, Methods and Evaluation

Abstract

Time series data is a collection of chronological observations which is generated by several domains such as medical and financial fields. Over the years, different tasks such as classification, forecasting, and clustering have been proposed to analyze this type of data. Time series data has been also used to study the effect of interventions over time. Moreover, in many fields of science, learning the causal structure of dynamic systems and time series data is considered an interesting task which plays an important role in scientific discoveries. Estimating the effect of an intervention and identifying the causal relations from the data can be performed via causal inference. Existing surveys on time series discuss traditional tasks such as classification and forecasting or explain the details of the approaches proposed to solve a specific task. In this paper, we focus on two causal inference tasks, i.e., treatment effect estimation and causal discovery for time series data, and provide a comprehensive review of the approaches in each task. Furthermore, we curate a list of commonly used evaluation metrics and datasets for each task and provide in-depth insight. These metrics and datasets can serve as benchmarks for research in the field.

References in corpus (36)

• Measuring Information Transfer
• Deep learning for time series classification: a review
• Granger causality and transfer entropy are equivalent for Gaussian variables
• Inferring causal impact using Bayesian structural time-series models
• C-RNN-GAN: Continuous recurrent neural networks with adversarial training
• Distinguishing cause from effect using observational data: methods and benchmarks
• Causal Effect Inference with Deep Latent-Variable Models
• Gaussian Process Kernels for Pattern Discovery and Extrapolation
• Causal Network Inference by Optimal Causation Entropy
• Kernel Interpolation for Scalable Structured Gaussian Processes (KISS-GP)
• Causation Entropy Identifies Indirect Influences, Dominance of Neighbors and Anticipatory Couplings
• Anomaly Detection in Univariate Time-series: A Survey on the State-of-the-Art
• Granger Causality in Multi-variate Time Series using a Time Ordered Restricted Vector Autoregressive Model
• Stochastic Variational Deep Kernel Learning
• Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Data
• Inference in Deep Gaussian Processes using Stochastic Gradient Hamiltonian Monte Carlo
• A scalable end-to-end Gaussian process adapter for irregularly sampled time series classification
• Time-series modeling with undecimated fully convolutional neural networks
• The Great Time Series Classification Bake Off: An Experimental Evaluation of Recently Proposed Algorithms. Extended Version
• Discovering contemporaneous and lagged causal relations in autocorrelated nonlinear time series datasets
• State Space LSTM Models with Particle MCMC Inference
• backShift: Learning causal cyclic graphs from unknown shift interventions
• Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions
• A Bayesian Approach to Learning Causal Networks
• Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models
• Deep Factors with Gaussian Processes for Forecasting
• Differentiable Causal Discovery from Interventional Data
• HyperST-Net: Hypernetworks for Spatio-Temporal Forecasting
• Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders
• seq2graph: Discovering Dynamic Dependencies from Multivariate Time Series with Multi-level Attention
• Economy Statistical Recurrent Units For Inferring Nonlinear Granger Causality
• Estimating heterogeneous treatment effects in nonstationary time series with state-space models
• Estimating Individual Treatment Effects with Time-Varying Confounders
• Estimating Granger Causality with Unobserved Confounders via Deep Latent-Variable Recurrent Neural Network
• Discovering Graphical Granger Causality Using the Truncating Lasso Penalty
• Learning Leading Indicators for Time Series Predictions