We address the problem of distribution shift in financial time series prediction, where the behavior of the time series changes over time.Satisfactory performance of forecasting algorithms requires constant model recalibration or fine-tuning to adapt to the new data distribution.Specifically, the ability to quickly fine-tune a model with only a few training samples available from the new distribution is crucial for many business applications.In this paper, we develop a novel method for learnable data augmentation that effectively adjusts to the new time series distribution with only a few samples. We demonstrate the effectiveness of our method compared to the state-of-the-art augmentation methods on both univariate time series (e.g., stock data) and multivariate time series (e.g., yield rate curves) in the presence of distribution shift due to the COVID market shock in 2020.
Dat Huynh (Northeastern University)
Hi, I am a Ph.D. candidate at Khoury College of Computer Sciences, Northeastern University, advised by Prof. Ehsan Elhamifar. I received my B.S. from University of Sciences (Viet Nam) where I was fortunate to study in Advanced Program in Computer Science.
Elizabeth Fons (J.P. Morgan Chase)
Svitlana Vyetrenko (J. P. Morgan, Artificial Intelligence Research)
More from the Same Authors
2021 : Efficient Calibration of Multi-Agent Market Simulators from Time Series with Bayesian Optimization »
Yuanlu Bai · Svitlana Vyetrenko · Henry Lam · Tucker Balch
2022 : A Synthetic Limit Order Book Dataset for Benchmarking Forecasting Algorithms under Distributional Shift »
Defu Cao · Yousef El-Laham · Loc Trinh · Svitlana Vyetrenko · Yan Liu
2022 : HyperTime: Implicit Neural Representations for Time Series »
Elizabeth Fons · Alejandro Sztrajman · Yousef El-Laham · Alexandros Iosifidis · Svitlana Vyetrenko
2020 Workshop: Fair AI in Finance »
Senthil Kumar · Cynthia Rudin · John Paisley · Isabelle Moulinier · C. Bayan Bruss · Eren K. · Susan Tibbs · Oluwatobi Olabiyi · Simona Gandrabur · Svitlana Vyetrenko · Kevin Compher