CNN-DRL with Shuffled Features in Finance ArXiv ID: 2402.03338 “View on arXiv” Authors: Unknown Abstract In prior methods, it was observed that the application of Convolutional Neural Networks agent in Deep Reinforcement Learning to financial data resulted in an enhanced reward. In this study, a specific permutation was applied to the feature vector, thereby generating a CNN matrix that strategically positions more pertinent features in close proximity. Our comprehensive experimental evaluations unequivocally demonstrate a substantial enhancement in reward attainment. ...

Do backrun auctions protect traders? ArXiv ID: 2401.08302 “View on arXiv” Authors: Unknown Abstract We study a new “laminated” queueing model for orders on batched trading venues such as decentralised exchanges. The model aims to capture and generalise transaction queueing infrastructure that has arisen to organise MEV activity on public blockchains such as Ethereum, providing convenient channels for sophisticated agents to extract value by acting on end-user order flow by performing arbitrage and related HFT activities. In our model, market orders are interspersed with orders created by arbitrageurs that under idealised conditions reset the marginal price to a global equilibrium between each trade, improving predictability of execution for liquidity traders. If an arbitrageur has a chance to land multiple opportunities in a row, he may attempt to manipulate the execution price of the intervening market order by a probabilistic blind sandwiching strategy. To study how bad this manipulation can get, we introduce and bound a price manipulation coefficient that measures the deviation from global equilibrium of local pricing quoted by a rational arbitrageur. We exhibit cases in which this coefficient is well approximated by a “zeta value’ with interpretable and empirically measurable parameters. ...

Dynamic portfolio selection under generalized disappointment aversion ArXiv ID: 2401.08323 “View on arXiv” Authors: Unknown Abstract This paper addresses the continuous-time portfolio selection problem under generalized disappointment aversion (GDA). The implicit definition of the certainty equivalent within GDA preferences introduces time inconsistency to this problem. We provide the sufficient and necessary condition for a strategy to be an equilibrium by a fully nonlinear integral equation. Investigating the existence and uniqueness of the solution to the integral equation, we establish the existence and uniqueness of the equilibrium. Our findings indicate that under disappointment aversion preferences, non-participation in the stock market is the unique equilibrium. The semi-analytical equilibrium strategies obtained under the constant relative risk aversion utility functions reveal that, under GDA preferences, the investment proportion in the stock market consistently remains smaller than the investment proportion under classical expected utility theory. The numerical analysis shows that the equilibrium strategy’s monotonicity concerning the two parameters of GDA preference aligns with the monotonicity of the degree of risk aversion. ...

Fitting random cash management models to data ArXiv ID: 2401.08548 “View on arXiv” Authors: Unknown Abstract Organizations use cash management models to control balances to both avoid overdrafts and obtain a profit from short-term investments. Most management models are based on control bounds which are derived from the assumption of a particular cash flow probability distribution. In this paper, we relax this strong assumption to fit cash management models to data by means of stochastic and linear programming. We also introduce ensembles of random cash management models which are built by randomly selecting a subsequence of the original cash flow data set. We illustrate our approach by means of a real case study showing that a small random sample of data is enough to fit sufficiently good bound-based models. ...

Forecasting Cryptocurrency Staking Rewards ArXiv ID: 2401.10931 “View on arXiv” Authors: Unknown Abstract This research explores a relatively unexplored area of predicting cryptocurrency staking rewards, offering potential insights to researchers and investors. We investigate two predictive methodologies: a) a straightforward sliding-window average, and b) linear regression models predicated on historical data. The findings reveal that ETH staking rewards can be forecasted with an RMSE within 0.7% and 1.1% of the mean value for 1-day and 7-day look-aheads respectively, using a 7-day sliding-window average approach. Additionally, we discern diverse prediction accuracies across various cryptocurrencies, including SOL, XTZ, ATOM, and MATIC. Linear regression is identified as superior to the moving-window average for perdicting in the short term for XTZ and ATOM. The results underscore the generally stable and predictable nature of staking rewards for most assets, with MATIC presenting a noteworthy exception. ...

Graph database while computationally efficient filters out quickly the ESG integrated equities in investment management ArXiv ID: 2401.07483 “View on arXiv” Authors: Unknown Abstract Design/methodology/approach This research evaluated the databases of SQL, No-SQL and graph databases to compare and contrast efficiency and performance. To perform this experiment the data were collected from multiple sources including stock price and financial news. Python is used as an interface to connect and query databases (to create database structures according to the feed file structure, to load data into tables, objects, to read data , to connect PostgreSQL, ElasticSearch, Neo4j. Purpose Modern applications of LLM (Large language model) including RAG (Retrieval Augmented Generation) with Machine Learning, deep learning, NLP (natural language processing) or Decision Analytics are computationally expensive. Finding a better option to consume less resources and time to get the result. Findings The Graph database of ESG (Environmental, Social and Governance) is comparatively better and can be considered for extended analytics to integrate ESG in business and investment. Practical implications A graph ML with a RAG architecture model can be introduced as a new framework with less computationally expensive LLM application in the equity filtering process for portfolio management. Originality/value Filtering out selective stocks out of two thousand or more listed companies in any stock exchange for active investment, consuming less resource consumption especially memory and energy to integrate artificial intelligence and ESG in business and investment. ...

Optimal Investment with Herd Behaviour Using Rational Decision Decomposition ArXiv ID: 2401.07183 “View on arXiv” Authors: Unknown Abstract In this paper, we study the optimal investment problem considering the herd behaviour between two agents, including one leading expert and one following agent whose decisions are influenced by those of the leading expert. In the objective functional of the optimal investment problem, we introduce the average deviation term to measure the distance between the two agents’ decisions and use the variational method to find its analytical solution. To theoretically analyze the impact of the following agent’s herd behaviour on his/her decision, we decompose his/her optimal decision into a convex linear combination of the two agents’ rational decisions, which we call the rational decision decomposition. Furthermore, we define the weight function in the rational decision decomposition as the following agent’s investment opinion to measure the preference of his/her own rational decision over that of the leading expert. We use the investment opinion to quantitatively analyze the impact of the herd behaviour, the following agent’s initial wealth, the excess return, and the volatility of the risky asset on the optimal decision. We validate our analyses through numerical experiments on real stock data. This study is crucial to understanding investors’ herd behaviour in decision-making and designing effective mechanisms to guide their decisions. ...

A deep implicit-explicit minimizing movement method for option pricing in jump-diffusion models ArXiv ID: 2401.06740 “View on arXiv” Authors: Unknown Abstract We develop a novel deep learning approach for pricing European basket options written on assets that follow jump-diffusion dynamics. The option pricing problem is formulated as a partial integro-differential equation, which is approximated via a new implicit-explicit minimizing movement time-stepping approach, involving approximation by deep, residual-type Artificial Neural Networks (ANNs) for each time step. The integral operator is discretized via two different approaches: (a) a sparse-grid Gauss-Hermite approximation following localised coordinate axes arising from singular value decompositions, and (b) an ANN-based high-dimensional special-purpose quadrature rule. Crucially, the proposed ANN is constructed to ensure the appropriate asymptotic behavior of the solution for large values of the underlyings and also leads to consistent outputs with respect to a priori known qualitative properties of the solution. The performance and robustness with respect to the dimension of these methods are assessed in a series of numerical experiments involving the Merton jump-diffusion model, while a comparison with the deep Galerkin method and the deep BSDE solver with jumps further supports the merits of the proposed approach. ...

Equity auction dynamics: latent liquidity models with activity acceleration ArXiv ID: 2401.06724 “View on arXiv” Authors: Unknown Abstract Equity auctions display several distinctive characteristics in contrast to continuous trading. As the auction time approaches, the rate of events accelerates causing a substantial liquidity buildup around the indicative price. This, in turn, results in a reduced price impact and decreased volatility of the indicative price. In this study, we adapt the latent/revealed order book framework to the specifics of equity auctions. We provide precise measurements of the model parameters, including order submissions, cancellations, and diffusion rates. Our setup allows us to describe the full dynamics of the average order book during closing auctions in Euronext Paris. These findings support the relevance of the latent liquidity framework in describing limit order book dynamics. Lastly, we analyze the factors contributing to a sub-diffusive indicative price and demonstrate the absence of indicative price predictability. ...

SpotV2Net: Multivariate Intraday Spot Volatility Forecasting via Vol-of-Vol-Informed Graph Attention Networks ArXiv ID: 2401.06249 “View on arXiv” Authors: Unknown Abstract This paper introduces SpotV2Net, a multivariate intraday spot volatility forecasting model based on a Graph Attention Network architecture. SpotV2Net represents assets as nodes within a graph and includes non-parametric high-frequency Fourier estimates of the spot volatility and co-volatility as node features. Further, it incorporates Fourier estimates of the spot volatility of volatility and co-volatility of volatility as features for node edges, to capture spillover effects. We test the forecasting accuracy of SpotV2Net in an extensive empirical exercise, conducted with the components of the Dow Jones Industrial Average index. The results we obtain suggest that SpotV2Net yields statistically significant gains in forecasting accuracy, for both single-step and multi-step forecasts, compared to a panel heterogenous auto-regressive model and alternative machine-learning models. To interpret the forecasts produced by SpotV2Net, we employ GNNExplainer \citep{“ying2019gnnexplainer”}, a model-agnostic interpretability tool, and thereby uncover subgraphs that are critical to a node’s predictions. ...