Identifiability and estimation of structural vector autoregressive models for subsampled and mixed-frequency time series.

@article{Tank2019IdentifiabilityAE,
  title={Identifiability and estimation of structural vector autoregressive models for subsampled and mixed-frequency time series.},
  author={Alex Tank and Emily B. Fox and Ali Shojaie},
  journal={Biometrika},
  year={2019},
  volume={106 2},
  pages={
          433-452
        }
}
Causal inference in multivariate time series is challenging because the sampling rate may not be as fast as the time scale of the causal interactions, so the observed series is a subsampled version of the desired series. Furthermore, series may be observed at different sampling rates, yielding mixed-frequency series. To determine instantaneous and lagged effects between series at the causal scale, we take a model-based approach that relies on structural vector autoregressive models. We present… 

Figures and Tables from this paper

Causal Discovery from Temporally Aggregated Time Series
TLDR
This paper investigates how to recover causal relations at the original causal frequency from temporally aggregated data when k is known and presents an estimation method based on non-Gaussian state-space modeling and evaluates its performance on both synthetic and real data.
Discrete Mixtures of Normals Pseudo Maximum Likelihood Estimators of Structural Vector Autoregressions
Likelihood inference in structural vector autoregressions with independent non-Gaussian shocks leads to parametric identification and efficient estimation at the risk of inconsistencies under
Granger Causality: A Review and Recent Advances
  • A. Shojaie, E. Fox
  • Computer Science
    Annual Review of Statistics and Its Application
  • 2021
TLDR
Recent advances that address various shortcomings of the earlier approaches are discussed, from models for high-dimensional time series to more recent developments that account for nonlinear and non-Gaussian observations and allow for subsampled and mixed-frequency time series.
Invariant Causal Prediction for Sequential Data
ABSTRACT We investigate the problem of inferring the causal predictors of a response Y from a set of d explanatory variables (X1, …, Xd). Classical ordinary least-square regression includes all
DYNOTEARS: Structure Learning from Time-Series Data
TLDR
This work revisits the structure learning problem for dynamic Bayesian networks and proposes a method that simultaneously estimates contemporaneous and time-lagged relationships between variables in a time-series, using a recent algebraic result characterizing the acyclicity constraint as a smooth equality constraint.
Likelihood-Free Overcomplete ICA and Applications in Causal Discovery
TLDR
A Likelihood-Free Overcomplete ICA algorithm (LFOICA) is presented that estimates the mixing matrix directly by back-propagation without any explicit assumptions on the density function of independent components, making a number of causal discovery procedures much more practically feasible.
A Lucas Critique Compliant SVAR model with Observation-driven Time-varying Parameters
We propose an observation-driven time-varying SVAR model where, in agreement with the Lucas Critique, structural shocks drive both the evolution of the macro variables and the dynamics of the VAR
Amalgamating evidence of dynamics
TLDR
This paper argues that amalgamation should proceed by integrating causal knowledge, rather than at the level of “raw” evidence, and defends this claim by first outlining both of these problems, and then showing that they can be solved only if the authors operate on causal structures.
Causal Discovery in High-Dimensional Point Process Networks with Hidden Nodes
TLDR
This work proposes a deconfounding procedure to estimate high-dimensional point process networks with only a subset of the nodes being observed, which allows flexible connections between the observed and unobserved processes.
A Nonconvex Framework for Structured Dynamic Covariance Recovery
TLDR
A two-stage optimization scheme with a carefully tailored spectral initialization, combined with iteratively refined alternating projected gradient descent is designed, which proves a linear convergence rate up to a nontrivial statistical error for the proposed descent scheme and establishes sample complexity guarantees for the estimator.
...
...

References

SHOWING 1-10 OF 47 REFERENCES
Causal Inference on Time Series using Restricted Structural Equation Models
TLDR
This work studies a class of restricted Structural Equation Models for time series that require independent residual time series, and shows empirically that when the data are causally insufficient or the model is misspecified, the method avoids incorrect answers.
Discovering Temporal Causal Relations from Subsampled Data
TLDR
This paper proposes an Expectation-Maximization approach and a variational inference approach to recover temporal causal relations from such subsampled data and proves that if the noise terms are non-Gaussian, the underlying model for the highfrequency data is identifiable from subsampling data under mild conditions.
Macroeconomics and the Reality of Mixed Frequency Data
Many time series are sampled at different frequencies. When we study co-movements between such series we usually analyze the joint process sampled at a common low frequency. This has consequences in
MULTIVARIATE AR SYSTEMS AND MIXED FREQUENCY DATA: G-IDENTIFIABILITY AND ESTIMATION
This paper is concerned with the problem of identifiability of the parameters of a high frequency multivariate autoregressive model from mixed frequency time series data. We demonstrate
Cointegration Analysis with Mixed-Frequency Data of Quarterly GDP and Monthly Coincident Indicators
Abstract The article introduces a method to estimate a cointegrated vector autoregressive model, using mixed-frequency data, in terms of a state-space representation of the vector error
Causal Discovery from Subsampled Time Series Data by Constraint Optimization
TLDR
A constraint satisfaction procedure is provided whose computational performance is several orders of magnitude better than previous approaches and allows for a robust and non-parametric estimation of system timescale causal structures from subsampled time series data.
Estimation of a Structural Vector Autoregression Model Using Non-Gaussianity
TLDR
This work shows how to combine the non-Gaussian instantaneous model with autoregressive models, effectively what is called a structural vector autoregression (SVAR) model, and contributes to the long-standing problem of how to estimate SVAR's.
New Introduction to Multiple Time Series Analysis
This reference work and graduate level textbook considers a wide range of models and methods for analyzing and forecasting multiple time series. The models covered include vector autoregressive,
Temporal aggregation and spurious instantaneous causality in multiple time series models
Large aggregation interval asymptotics are used to investigate the relation between Granger causality in disaggregated vector autoregressions (VARs) and associated contemporaneous correlation among
...
...