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/

Zendy Plus

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

Zendy Tools

Zendy Open

Model-Free Reinforcement Learning has achieved meaningful results in stableenvironments but, to this day, it remains problematic in regime changingenvironments like financial markets. In contrast, model-based RL is able tocapture some fundamental and dynamical concepts of the environment but sufferfrom cognitive bias. In this work, we propose to combine the best of the twotechniques by selecting various model-based approaches thanks to Model-FreeDeep Reinforcement Learning. Using not only past performance and volatility, weinclude additional contextual information such as macro and risk appetitesignals to account for implicit regime changes. We also adapt traditional RLmethods to real-life situations by considering only past data for the trainingsets. Hence, we cannot use future information in our training data set asimplied by K-fold cross validation. Building on traditional statisticalmethods, we use the traditional "walk-forward analysis", which is defined bysuccessive training and testing based on expanding periods, to assert therobustness of the resulting agent.  Finally, we present the concept of statistical difference's significancebased on a two-tailed T-test, to highlight the ways in which our models differfrom more traditional ones. Our experimental results show that our approachoutperforms traditional financial baseline portfolio models such as theMarkowitz model in almost all evaluation metrics commonly used in financialmathematics, namely net performance, Sharpe and Sortino ratios, maximumdrawdown, maximum drawdown over volatility.

Adaptive learning for financial markets mixing model-based and model-free RL for volatility targeting