arXiv subjects

Chung I Lu

Publications and source records attributed to Chung I Lu.

3 recordsLinked to original sources

arXiv2025

Generative modelling of financial time series with structured noise and MMD-based signature learning

Generating synthetic financial time series data that accurately reflects real-world market dynamics holds tremendous potential for various applications, including portfolio optimization, risk management, and large scale machine learning. We present an approach that {uses structured noise} for training generative models for financial time series. The expressive power of the signature transform {has been shown to be able} to capture the complex dependencies and temporal structures inherent in financial data {when used to train generative models in the form of a signature kernel }. We employ a moving average model to model the variance of the noise input, enhancing the model's ability to reproduce stylized facts such as volatility clustering. Through empirical experiments on S\&P 500 index data, we demonstrate that our model effectively captures key characteristics of financial time series and outperforms comparable {approaches}. In addition, we explore the application of the synthetic data generated to train a reinforcement learning agent for portfolio management, achieving promising results. Finally, we propose a method to add robustness to the generative model by tweaking the noise input so that the generated sequences can be adjusted to different market environments with minimal data.

q-fin.MF↗

arXiv2025

Evaluation of Deep Reinforcement Learning Algorithms for Portfolio Optimisation

We evaluate benchmark deep reinforcement learning algorithms on the task of portfolio optimisation using simulated data. The simulator to generate the data is based on correlated geometric Brownian motion with the Bertsimas-Lo market impact model. Using the Kelly criterion (log utility) as the objective, we can analytically derive the optimal policy without market impact as an upper bound to measure performance when including market impact. We find that the off-policy algorithms DDPG, TD3 and SAC are unable to learn the right $Q$-function due to the noisy rewards and therefore perform poorly. The on-policy algorithms PPO and A2C, with the use of generalised advantage estimation, are able to deal with the noise and derive a close to optimal policy. The clipping variant of PPO was found to be important in preventing the policy from deviating from the optimal once converged. In a more challenging environment where we have regime changes in the GBM parameters, we find that PPO, combined with a hidden Markov model to learn and predict the regime context, is able to learn different policies adapted to each regime. Overall, we find that the sample complexity of these algorithms is too high for applications using real data, requiring more than 2m steps to learn a good policy in the simplest setting, which is equivalent to almost 8,000 years of daily prices.

cs.CE↗

arXiv2025

Distributionally Robust Deep Q-Learning

We propose a novel distributionally robust $Q$-learning algorithm for the non-tabular case accounting for continuous state spaces where the state transition of the underlying Markov decision process is subject to model uncertainty. The uncertainty is taken into account by considering the worst-case transition from a ball around a reference probability measure. To determine the optimal policy under the worst-case state transition, we solve the associated non-linear Bellman equation by dualising and regularising the Bellman operator with the Sinkhorn distance, which is then parameterized with deep neural networks. This approach allows us to modify the Deep Q-Network algorithm to optimise for the worst case state transition. We illustrate the tractability and effectiveness of our approach through several applications, including a portfolio optimisation task based on S\&{P}~500 data.

cs.LG↗