Quantitative stock portfolio optimization by multi-task learning risk and return
Selecting profitable stocks for investments is a challenging task. Recent research has made significant progress on stock ranking prediction to select top-ranked stocks for portfolio optimization. However, the stocks are only ranked by predicted stock return, ignoring the stock price volatility risk...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/173235 |
| _version_ | 1826114886079348736 |
|---|---|
| author | Ma, Yu Mao, Rui Lin, Qika Wu, Peng Cambria, Erik |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Ma, Yu Mao, Rui Lin, Qika Wu, Peng Cambria, Erik |
| author_sort | Ma, Yu |
| collection | NTU |
| description | Selecting profitable stocks for investments is a challenging task. Recent research has made significant progress on stock ranking prediction to select top-ranked stocks for portfolio optimization. However, the stocks are only ranked by predicted stock return, ignoring the stock price volatility risk — a critical aspect for stock selection and investments. Moreover, they preliminarily attempted to capture the effects of related stocks from a singular relation, disregarding the rich information regarding multiple spillover effects from related stocks and the distinctions in effects among various relations. Thus, we propose a risk and return multi-task learning model with a heterogeneous graph attention network (HGA-MT) to predict stock ranking for portfolio optimization. First, to aggregate the multiple spillover effects of related stocks, we introduce graph convolutional networks to fuse the effects of related stocks in each relation and design an attention network to allocate varying weights to different types of relationships. Second, we use a multi-task learning paradigm to learn stock return and volatility risks jointly. The stock ranking results are calculated by simultaneously considering the risk and return. Thus, Top-K ranked stocks are recommended in the portfolio for the next trading day to achieve higher and more stable profits. Extensive experiments prove that HGA-MT outperforms previous state-of-the-art methods in stock ranking and backtesting trading evaluation tasks. |
| first_indexed | 2024-10-01T03:46:31Z |
| format | Journal Article |
| id | ntu-10356/173235 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:46:31Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1732352024-01-22T00:49:41Z Quantitative stock portfolio optimization by multi-task learning risk and return Ma, Yu Mao, Rui Lin, Qika Wu, Peng Cambria, Erik School of Computer Science and Engineering Engineering::Computer science and engineering Stock Portfolio Optimization Quantitative Stock Selection Selecting profitable stocks for investments is a challenging task. Recent research has made significant progress on stock ranking prediction to select top-ranked stocks for portfolio optimization. However, the stocks are only ranked by predicted stock return, ignoring the stock price volatility risk — a critical aspect for stock selection and investments. Moreover, they preliminarily attempted to capture the effects of related stocks from a singular relation, disregarding the rich information regarding multiple spillover effects from related stocks and the distinctions in effects among various relations. Thus, we propose a risk and return multi-task learning model with a heterogeneous graph attention network (HGA-MT) to predict stock ranking for portfolio optimization. First, to aggregate the multiple spillover effects of related stocks, we introduce graph convolutional networks to fuse the effects of related stocks in each relation and design an attention network to allocate varying weights to different types of relationships. Second, we use a multi-task learning paradigm to learn stock return and volatility risks jointly. The stock ranking results are calculated by simultaneously considering the risk and return. Thus, Top-K ranked stocks are recommended in the portfolio for the next trading day to achieve higher and more stable profits. Extensive experiments prove that HGA-MT outperforms previous state-of-the-art methods in stock ranking and backtesting trading evaluation tasks. Published version This paper was supported by the National Natural Science Foundation of China (project numbers are 72274096, 72174087, 71774084 and 71874082), the National Social Science Fund of China (project number is 17ZDA291), Foreign Cultural and Educational Expert Program of the Ministry of Science and Technology of China (G202218- 2009L). 2024-01-22T00:49:41Z 2024-01-22T00:49:41Z 2024 Journal Article Ma, Y., Mao, R., Lin, Q., Wu, P. & Cambria, E. (2024). Quantitative stock portfolio optimization by multi-task learning risk and return. Information Fusion, 104, 102165-. https://dx.doi.org/10.1016/j.inffus.2023.102165 1566-2535 https://hdl.handle.net/10356/173235 10.1016/j.inffus.2023.102165 2-s2.0-85178667497 104 102165 en Information Fusion © 2023 Elsevier B.V. All rights reserved. |
| spellingShingle | Engineering::Computer science and engineering Stock Portfolio Optimization Quantitative Stock Selection Ma, Yu Mao, Rui Lin, Qika Wu, Peng Cambria, Erik Quantitative stock portfolio optimization by multi-task learning risk and return |
| title | Quantitative stock portfolio optimization by multi-task learning risk and return |
| title_full | Quantitative stock portfolio optimization by multi-task learning risk and return |
| title_fullStr | Quantitative stock portfolio optimization by multi-task learning risk and return |
| title_full_unstemmed | Quantitative stock portfolio optimization by multi-task learning risk and return |
| title_short | Quantitative stock portfolio optimization by multi-task learning risk and return |
| title_sort | quantitative stock portfolio optimization by multi task learning risk and return |
| topic | Engineering::Computer science and engineering Stock Portfolio Optimization Quantitative Stock Selection |
| url | https://hdl.handle.net/10356/173235 |
| work_keys_str_mv | AT mayu quantitativestockportfoliooptimizationbymultitasklearningriskandreturn AT maorui quantitativestockportfoliooptimizationbymultitasklearningriskandreturn AT linqika quantitativestockportfoliooptimizationbymultitasklearningriskandreturn AT wupeng quantitativestockportfoliooptimizationbymultitasklearningriskandreturn AT cambriaerik quantitativestockportfoliooptimizationbymultitasklearningriskandreturn |