Modeling financial uncertainty with multivariate temporal entropy-based curriculums

Ramit Sawhney, Arnav Wadhwa, Ayush Mangal, Vivek Mittal, Shivam Agarwal, Rajiv Ratn Shah
Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, PMLR 161:1671-1681, 2021.

Abstract

In the financial realm, profit generation greatly relies on the complicated task of stock prediction. Lately, neural methods have shown success in exploiting stock affecting signals from textual data across news and tweets to forecast stock performance. However, the dynamic, stochastic, and variably influential nature of text and prices makes it difficult to train neural stock trading models, limiting predictive performance and profits. To transcend this limitation, we propose a novel multi-modal curriculum learning approach: FinCLASS, which evaluates stock affecting signals via entropy-based heuristics and measures their linguistic and price-based complexities in a time-aware, hierarchical fashion. We show that training financial models can benefit by exposing neural networks to easier examples of stock affecting signals early during the training phase, before introducing samples having more complex linguistic and price-based temporal variations. Through experiments on benchmark English tweets and Chinese financial news spanning two major indexes and four global markets, we show how FinCLASS outperforms state-of-the-art across financial tasks of stock movement prediction, volatility regression, and profit generation. Through ablative and qualitative experiments, we set the case for FinCLASS as a generalizable framework for developing natural language-centric neural models for financial tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v161-sawhney21a, title = {Modeling financial uncertainty with multivariate temporal entropy-based curriculums}, author = {Sawhney, Ramit and Wadhwa, Arnav and Mangal, Ayush and Mittal, Vivek and Agarwal, Shivam and Shah, Rajiv Ratn}, booktitle = {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence}, pages = {1671--1681}, year = {2021}, editor = {de Campos, Cassio and Maathuis, Marloes H.}, volume = {161}, series = {Proceedings of Machine Learning Research}, month = {27--30 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v161/sawhney21a/sawhney21a.pdf}, url = {https://proceedings.mlr.press/v161/sawhney21a.html}, abstract = {In the financial realm, profit generation greatly relies on the complicated task of stock prediction. Lately, neural methods have shown success in exploiting stock affecting signals from textual data across news and tweets to forecast stock performance. However, the dynamic, stochastic, and variably influential nature of text and prices makes it difficult to train neural stock trading models, limiting predictive performance and profits. To transcend this limitation, we propose a novel multi-modal curriculum learning approach: FinCLASS, which evaluates stock affecting signals via entropy-based heuristics and measures their linguistic and price-based complexities in a time-aware, hierarchical fashion. We show that training financial models can benefit by exposing neural networks to easier examples of stock affecting signals early during the training phase, before introducing samples having more complex linguistic and price-based temporal variations. Through experiments on benchmark English tweets and Chinese financial news spanning two major indexes and four global markets, we show how FinCLASS outperforms state-of-the-art across financial tasks of stock movement prediction, volatility regression, and profit generation. Through ablative and qualitative experiments, we set the case for FinCLASS as a generalizable framework for developing natural language-centric neural models for financial tasks.} }
Endnote
%0 Conference Paper %T Modeling financial uncertainty with multivariate temporal entropy-based curriculums %A Ramit Sawhney %A Arnav Wadhwa %A Ayush Mangal %A Vivek Mittal %A Shivam Agarwal %A Rajiv Ratn Shah %B Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2021 %E Cassio de Campos %E Marloes H. Maathuis %F pmlr-v161-sawhney21a %I PMLR %P 1671--1681 %U https://proceedings.mlr.press/v161/sawhney21a.html %V 161 %X In the financial realm, profit generation greatly relies on the complicated task of stock prediction. Lately, neural methods have shown success in exploiting stock affecting signals from textual data across news and tweets to forecast stock performance. However, the dynamic, stochastic, and variably influential nature of text and prices makes it difficult to train neural stock trading models, limiting predictive performance and profits. To transcend this limitation, we propose a novel multi-modal curriculum learning approach: FinCLASS, which evaluates stock affecting signals via entropy-based heuristics and measures their linguistic and price-based complexities in a time-aware, hierarchical fashion. We show that training financial models can benefit by exposing neural networks to easier examples of stock affecting signals early during the training phase, before introducing samples having more complex linguistic and price-based temporal variations. Through experiments on benchmark English tweets and Chinese financial news spanning two major indexes and four global markets, we show how FinCLASS outperforms state-of-the-art across financial tasks of stock movement prediction, volatility regression, and profit generation. Through ablative and qualitative experiments, we set the case for FinCLASS as a generalizable framework for developing natural language-centric neural models for financial tasks.
APA
Sawhney, R., Wadhwa, A., Mangal, A., Mittal, V., Agarwal, S. & Shah, R.R.. (2021). Modeling financial uncertainty with multivariate temporal entropy-based curriculums. Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 161:1671-1681 Available from https://proceedings.mlr.press/v161/sawhney21a.html.

Related Material