RNN-based counterfactual time-series prediction
Explore this paper's citation graph
Summary
Recurrent neural networks are used to predict counterfactual time-series of treated unit outcomes using only the outcomes of control units as inputs, and RNNs outperform SCM in terms of recovering experimental estimates from a field experiment extended to a time- series observational setting.
- Type
- article
- Published
- 2018-05-01
- Cited by
- 5
- References
- 28
- Access
- Open access
- OpenAlex
- https://openalex.org/W2795819716
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:126325258
Keywords
Counterfactual thinking, Recurrent neural network, Exploit, Computer science, Representation (politics)
References
- Attention-Based Models for Speech Recognition
- Field Experiments Testing the Impact of Radio Advertisements on Electoral Competition
- Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
- Long Short-Term Memory
- Inferring causal impact using Bayesian structural time-series models
- Hierarchical Recurrent Neural Networks for Long-Term Dependencies
- Comparative Politics and the Synthetic Control Method
- Do Political Parties Matter? Evidence from U.S. Cities-super-*
- Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California’s Tobacco Control Program
- Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
- The Economic Costs of Conflict: A Case Study of the
- Measuring Ad Effectiveness Using Geo Experiments
- Cherry Picking with Synthetic Controls
- Retrospective Voting in Big-City US Mayoral Elections
- Deep and Confident Prediction for Time Series at Uber
- Synthetic Control Methods: Never Use All Pre-Intervention Outcomes Together With Covariates
- Catastrophic Natural Disasters and Economic Growth
- Comparative politics and the synthetic control method revisited: a note on Abadie et al. (2015)
- Deep Learning
- Position-Based Content Attention for Time Series Forecasting with Sequence-to-Sequence RNNs
Cited by
- State-Building through Public Land Disposal? An Application of Matrix Completion for Counterfactual Prediction
- A Machine Learning Evaluation of the Effects of South Africa's COVID-19 Lockdown Measures on Population Mobility
- Estimating the Impact of an Improvement to a Revenue Management System: An Airline Application
- Assessing the Impact: Does an Improvement to a Revenue Management System Lead to an Improved Revenue?
- Two ways towards combining Sequential Neural Network and Statistical Methods to Improve the Prediction of Time Series
Related papers
- Counterfactual time-series prediction with encoder-decoder networks
- RECOWNs: Probabilistic Circuits for Trustworthy Time Series Forecasting
- Perfect Match: A Simple Method for Learning Representations For Counterfactual Inference With Neural Networks
- Estimating the Effects of Continuous-valued Interventions using Generative Adversarial Networks
- Counterfactual domain adversarial training of neural networks
- Teaching deep learning causal effects improves predictive performance
- Dynamical Systems Theory for Causal Inference with Application to Synthetic Control Methods