Causal Discovery in High-Dimensional Point Process Networks with Hidden Nodes
arXiv:2109.10947 · doi:10.3390/e23121622
Abstract
Thanks to technological advances leading to near-continuous time observations, emerging multivariate point process data offer new opportunities for causal discovery. However, a key obstacle in achieving this goal is that many relevant processes may not be observed in practice. Naive estimation approaches that ignore these hidden variables can generate misleading results because of the unadjusted confounding. To plug this gap, we propose a deconfounding procedure to estimate high-dimensional point process networks with only a subset of the nodes being observed. Our method allows flexible connections between the observed and unobserved processes. It also allows the number of unobserved processes to be unknown and potentially larger than the number of observed nodes. Theoretical analyses and numerical studies highlight the advantages of the proposed method in identifying causal interactions among the observed processes.
arXiv comment31 pages, 5 figures
References in corpus (12)
- Simultaneous whole-animal 3D-imaging of neuronal activity using light field microscopy
- Restricted strong convexity and weighted matrix completion: Optimal bounds with noise
- Regularized estimation in sparse high-dimensional time series models
- Statistical significance in high-dimensional linear models
- Lasso and probabilistic inequalities for multivariate point processes
- Discovering Latent Network Structure in Point Process Data
- Renewal in Hawkes processes with self-excitation and inhibition
- The Multivariate Hawkes Process in High Dimensions: Beyond Mutual Excitation
- A shotgun sampling solution for the common input problem in neural connectivity inference
- Effects of hidden nodes on network structure inference
- Statistical Inference for Networks of High-Dimensional Point Processes
- Causal Structural Learning Via Local Graphs