Weighted Tensor Completion for Time-Series Causal Inference

Marginal Structural Models (MSM) {Robins, 2000} are the most popular models for causal inference from time-series observational data. However, they have two main drawbacks: (a) they do not capture subject heterogeneity, and (b) they only consider fixed time intervals and do not scale gracefully with longer intervals. In this work, we propose a new family of MSMs to address these two concerns. We model the potential outcomes as a three-dimensional tensor of low rank, where the three dimensions correspond to the agents, time periods and the set of possible histories. Unlike the traditional MSM, we allow the dimensions of the tensor to increase with the number of agents and time periods. We set up a weighted tensor completion problem as our estimation procedure, and show that the solution to this problem converges to the true model in an appropriate sense. Then we show how to solve the estimation problem, providing conditions under which we can approximately and efficiently solve the estimation problem. Finally, we propose an algorithm based on projected gradient descent, which is easy to implement and evaluate its performance on a simulated dataset.

[1]  Adam Glynn,et al.  How to Make Causal Inferences with Time-Series Cross-Sectional Data * , 2013 .

[2]  Maja Pantic,et al.  TensorLy: Tensor Learning in Python , 2016, J. Mach. Learn. Res..

[3]  David P. Woodruff,et al.  Weighted low rank approximations with provable guarantees , 2016, STOC.

[4]  Sham M. Kakade,et al.  Learning mixtures of spherical gaussians: moment methods and spectral decompositions , 2012, ITCS '13.

[5]  Mihaela van der Schaar,et al.  Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders , 2019, ICML.

[6]  James M. Robins,et al.  Marginal Structural Models versus Structural nested Models as Tools for Causal inference , 2000 .

[7]  Uri Shalit,et al.  Learning Representations for Counterfactual Inference , 2016, ICML.

[8]  Tommi S. Jaakkola,et al.  Weighted Low-Rank Approximations , 2003, ICML.

[9]  Ankur Moitra,et al.  Noisy tensor completion via the sum-of-squares hierarchy , 2015, Mathematical Programming.

[10]  Romain Neugebauer,et al.  Causal inference in longitudinal studies with history-restricted marginal structural models. , 2007, Electronic journal of statistics.

[11]  Anima Anandkumar,et al.  A Spectral Algorithm for Latent Dirichlet Allocation , 2012, Algorithmica.

[12]  Uri Shalit,et al.  Estimating individual treatment effect: generalization bounds and algorithms , 2016, ICML.

[13]  Susan Athey,et al.  Recursive partitioning for heterogeneous causal effects , 2015, Proceedings of the National Academy of Sciences.

[14]  J. Robins.  A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , 1986 .

[15]  David P. Woodruff,et al.  Relative Error Tensor Low Rank Approximation , 2017, Electron. Colloquium Comput. Complex..

[16]  Andrea Montanari,et al.  Spectral Algorithms for Tensor Completion , 2016, ArXiv.

[17]  Suchi Saria,et al.  A Non-parametric Bayesian Approach for Estimating Treatment-Response Curves from Sparse Time Series , 2016, MLHC.

[18]  Jennifer L. Hill,et al.  Bayesian Nonparametric Modeling for Causal Inference , 2011 .

[19]  Ming Yuan,et al.  Non-Convex Projected Gradient Descent for Generalized Low-Rank Tensor Regression , 2016, J. Mach. Learn. Res..

[20]  J. Pearl.  Detecting Latent Heterogeneity , 2013, Probabilistic and Causal Inference.

[21]  K. Imai,et al.  Robust Estimation of Inverse Probability Weights for Marginal Structural Models , 2015 .

[22]  Ming Yuan,et al.  On Tensor Completion via Nuclear Norm Minimization , 2014, Foundations of Computational Mathematics.

[23]  Tyler J VanderWeele,et al.  A marginal structural model analysis for loneliness: implications for intervention trials and clinical practice. , 2011, Journal of consulting and clinical psychology.

[24]  J. Robins,et al.  Marginal Structural Models and Causal Inference in Epidemiology , 2000, Epidemiology.

[25]  Dario Bini.  Border rank of m×n×(mn−q) tensors , 1986 .

[26]  W. Newey,et al.  Large sample estimation and hypothesis testing , 1986 .

[27]  Mihaela van der Schaar,et al.  GANITE: Estimation of Individualized Treatment Effects using Generative Adversarial Nets , 2018, ICLR.

[28]  J. Robins,et al.  Doubly Robust Estimation in Missing Data and Causal Inference Models , 2005, Biometrics.

[29]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[30]  Neil Shephard,et al.  Time Series Experiments and Causal Estimands: Exact Randomization Tests and Trading , 2017 .

[31]  Bernhard Schölkopf,et al.  Causal Inference on Time Series using Restricted Structural Equation Models , 2013, NIPS.

[32]  Judea Pearl,et al.  Identification of Conditional Interventional Distributions , 2006, UAI.

[33]  Daron Acemoglu,et al.  Democracy Does Cause Growth , 2014, Journal of Political Economy.

[34]  Stefan Wager,et al.  Estimation and Inference of Heterogeneous Treatment Effects using Random Forests , 2015, Journal of the American Statistical Association.

[35]  S. Murphy,et al.  Assessing Time-Varying Causal Effect Moderation in Mobile Health , 2016, Journal of the American Statistical Association.

[36]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[37]  Bryan Lim,et al.  Forecasting Treatment Responses Over Time Using Recurrent Marginal Structural Networks , 2018, NeurIPS.

[38]  G. Imbens,et al.  Matrix Completion Methods for Causal Panel Data Models , 2017, Journal of the American Statistical Association.

[39]  J. Pearl,et al.  The Book of Why: The New Science of Cause and Effect , 2018 .

[40]  Suchi Saria,et al.  Learning Treatment-Response Models from Multivariate Longitudinal Data , 2017, UAI.