English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

With the fast development of quantitative portfolio optimization in financialengineering, lots of AI-based algorithmic trading strategies have demonstratedpromising results, among which reinforcement learning begins to manifestcompetitive advantages. However, the environment from real financial markets iscomplex and hard to be fully simulated, considering the observation of abrupttransitions, unpredictable hidden causal factors, heavy tail properties and soon. Thus, in this paper, first, we adopt a heavy-tailed preserving normalizingflows to simulate high-dimensional joint probability of the complex tradingenvironment and develop a model-based reinforcement learning framework tobetter understand the intrinsic mechanisms of quantitative online trading.Second, we experiment with various stocks from three different financialmarkets (Dow, NASDAQ and S&P) and show that among these three financialmarkets, Dow gets the best performance based on various evaluation metricsunder our back-testing system. Especially, our proposed method is able tomitigate the impact of unpredictable financial market crises during theCOVID-19 pandemic period, resulting in a lower maximum drawdown. Third, we alsoexplore the explanation of our RL algorithm. (1), we utilize the patterncausality method to study the interactive relation among different stocks inthe environment. (2), We analyze the dynamic loss and actor loss to ensure theconvergence of our strategies. (3), by visualizing high dimensional statetransition data comparisons from real and virtual buffers with t-SNE, weuncover some effective patterns of better portfolio optimization strategies.(4), we also utilize eigenvalue analysis to study the convergence properties ofthe environmen's model.

Model Based Reinforcement Learning with Non-Gaussian Environment Dynamics and its Application to Portfolio Optimization