Deep deterministic portfolio optimization
| Autor: | Stephen J. Hardiman, Joachim de Lataillade, Christian Schmidt, Emmanuel Sérié, Ayman Chaouki |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Statistics and Probability Computer Science - Machine Learning Economics and Econometrics Mathematical optimization Computer science lcsh:QA75.5-76.95 Machine Learning (cs.LG) FOS: Economics and business Simple (abstract algebra) 0502 economics and business Reinforcement learning lcsh:Finance lcsh:HG1-9999 Trading strategy 050207 economics 050208 finance Applied Mathematics Portfolio optimization 05 social sciences Mathematical Finance (q-fin.MF) Computer Science Applications Work (electrical) Quantitative Finance - Mathematical Finance Stochastic control Business Management and Accounting (miscellaneous) lcsh:Electronic computers. Computer science Finance |
| Zdroj: | Journal of Finance and Data Science, Vol 6, Iss, Pp 16-30 (2020) |
| ISSN: | 2405-9188 |
| Popis: | Can deep reinforcement learning algorithms be exploited as solvers for optimal trading strategies? The aim of this work is to test reinforcement learning algorithms on conceptually simple, but mathematically non-trivial, trading environments. The environments are chosen such that an optimal or close-to-optimal trading strategy is known. We study the deep deterministic policy gradient algorithm and show that such a reinforcement learning agent can successfully recover the essential features of the optimal trading strategies and achieve close-to-optimal rewards. Comment: Minor typo |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|