MSPM(b) outperforms all baselines on Portfolio(b) in terms of the accumulated portfolio value in back-testing.

<p>MSPM(b) outperforms all baselines on Portfolio(b) in terms of the accumulated portfolio value in back-testing.</p>

Saved in:

Bibliographic Details
Main Author:	Zhenhan Huang (12107002) (author)
Other Authors:	Fumihide Tanaka (12107005) (author)
Published:	2022
Subjects:	Biotechnology Science Policy Biological Sciences not elsewhere classified Information Systems not elsewhere classified receives heterogeneous data heterogeneous data input five different baselines proximal policy optimization constant rebalanced portfolio mspm simultaneously addresses mspm improves arr volatile market information used pm strategy strategic agent module evolving agent module four different portfolios agent reinforcement learning reinforcement learning portfolio optimization different portfolios making module generating module financial portfolio mspm ). agent rl year u varying number updated modules two challenges step condensation sr ). sortino ratio reusable design reusable asset profit accumulation particular asset modular design making nature least 186 least 1341 increasing need hardly reusable generates signal drr ), deep q dedicated agents crp ), comprised information compare mspms changing markets based system based pm accumulated rate
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items