Deconvolutional Time Series Regression: A Technique for Modeling Temporally Diffuse Effects

Cory Shain, William Schuler


Abstract
Researchers in computational psycholinguistics frequently use linear models to study time series data generated by human subjects. However, time series may violate the assumptions of these models through temporal diffusion, where stimulus presentation has a lingering influence on the response as the rest of the experiment unfolds. This paper proposes a new statistical model that borrows from digital signal processing by recasting the predictors and response as convolutionally-related signals, using recent advances in machine learning to fit latent impulse response functions (IRFs) of arbitrary shape. A synthetic experiment shows successful recovery of true latent IRFs, and psycholinguistic experiments reveal plausible, replicable, and fine-grained estimates of latent temporal dynamics, with comparable or improved prediction quality to widely-used alternatives.
Anthology ID:
D18-1288
Volume:
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
Month:
October-November
Year:
2018
Address:
Brussels, Belgium
Editors:
Ellen Riloff, David Chiang, Julia Hockenmaier, Jun’ichi Tsujii
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2679–2689
Language:
URL:
https://aclanthology.org/D18-1288
DOI:
10.18653/v1/D18-1288
Bibkey:
Cite (ACL):
Cory Shain and William Schuler. 2018. Deconvolutional Time Series Regression: A Technique for Modeling Temporally Diffuse Effects. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2679–2689, Brussels, Belgium. Association for Computational Linguistics.
Cite (Informal):
Deconvolutional Time Series Regression: A Technique for Modeling Temporally Diffuse Effects (Shain & Schuler, EMNLP 2018)
Copy Citation:
PDF:
https://aclanthology.org/D18-1288.pdf
Code
 coryshain/dtsr
Data
Natural Stories