Kronos: A Foundation Model for the Language of Financial Markets ArXiv ID: 2508.02739 “View on arXiv” Authors: Yu Shi, Zongliang Fu, Shuo Chen, Bohan Zhao, Wei Xu, Changshui Zhang, Jian Li Abstract The success of large-scale pre-training paradigm, exemplified by Large Language Models (LLMs), has inspired the development of Time Series Foundation Models (TSFMs). However, their application to financial candlestick (K-line) data remains limited, often underperforming non-pre-trained architectures. Moreover, existing TSFMs often overlook crucial downstream tasks such as volatility prediction and synthetic data generation. To address these limitations, we propose Kronos, a unified, scalable pre-training framework tailored to financial K-line modeling. Kronos introduces a specialized tokenizer that discretizes continuous market information into token sequences, preserving both price dynamics and trade activity patterns. We pre-train Kronos using an autoregressive objective on a massive, multi-market corpus of over 12 billion K-line records from 45 global exchanges, enabling it to learn nuanced temporal and cross-asset representations. Kronos excels in a zero-shot setting across a diverse set of financial tasks. On benchmark datasets, Kronos boosts price series forecasting RankIC by 93% over the leading TSFM and 87% over the best non-pre-trained baseline. It also achieves a 9% lower MAE in volatility forecasting and a 22% improvement in generative fidelity for synthetic K-line sequences. These results establish Kronos as a robust, versatile foundation model for end-to-end financial time series analysis. Our pre-trained model is publicly available at https://github.com/shiyu-coder/Kronos. ...

Two Stochastic Control Methods for Mean-Variance Portfolio Selection of Jump Diffusions and Their Relationship ArXiv ID: 2508.01138 “View on arXiv” Authors: Qiyue Zhang, Jingtao Shi Abstract This paper is concerned with the maximum principle and dynamic programming principle for mean-variance portfolio selection of jump diffusions and their relationship. First, the optimal portfolio and efficient frontier of the problem are obtained using both methods. Furthermore, the relationship between these two methods is investigated. Specially, the connections between the adjoint processes and value function are given. ...

ContestTrade: A Multi-Agent Trading System Based on Internal Contest Mechanism ArXiv ID: 2508.00554 “View on arXiv” Authors: Li Zhao, Rui Sun, Zuoyou Jiang, Bo Yang, Yuxiao Bai, Mengting Chen, Xinyang Wang, Jing Li, Zuo Bai Abstract In financial trading, large language model (LLM)-based agents demonstrate significant potential. However, the high sensitivity to market noise undermines the performance of LLM-based trading systems. To address this limitation, we propose a novel multi-agent system featuring an internal competitive mechanism inspired by modern corporate management structures. The system consists of two specialized teams: (1) Data Team - responsible for processing and condensing massive market data into diversified text factors, ensuring they fit the model’s constrained context. (2) Research Team - tasked with making parallelized multipath trading decisions based on deep research methods. The core innovation lies in implementing a real-time evaluation and ranking mechanism within each team, driven by authentic market feedback. Each agent’s performance undergoes continuous scoring and ranking, with only outputs from top-performing agents being adopted. The design enables the system to adaptively adjust to dynamic environment, enhances robustness against market noise and ultimately delivers superior trading performance. Experimental results demonstrate that our proposed system significantly outperforms prevailing multi-agent systems and traditional quantitative investment methods across diverse evaluation metrics. ContestTrade is open-sourced on GitHub at https://github.com/FinStep-AI/ContestTrade. ...

Complexity of Financial Time Series: Multifractal and Multiscale Entropy Analyses ArXiv ID: 2507.23414 “View on arXiv” Authors: Oday Masoudi, Farhad Shahbazi, Mohammad Sharifi Abstract We employed Multifractal Detrended Fluctuation Analysis (MF-DFA) and Refined Composite Multiscale Sample Entropy (RCMSE) to investigate the complexity of Bitcoin, GBP/USD, gold, and natural gas price log-return time series. This study provides a comparative analysis of these markets and offers insights into their predictability and associated risks. Each tool presents a unique method to quantify time series complexity. The RCMSE and MF-DFA methods demonstrate a higher complexity for the Bitcoin time series than others. It is discussed that the increased complexity of Bitcoin may be attributable to the presence of higher nonlinear correlations within its log-return time series. ...

Is Causality Necessary for Efficient Portfolios? A Computational Perspective on Predictive Validity and Model Misspecification ArXiv ID: 2507.23138 “View on arXiv” Authors: Alejandro Rodriguez Dominguez Abstract A recent line of research has argued that causal factor models are necessary for portfolio optimization, claiming that structurally misspecified models inevitably produce inverted signals and nonviable frontiers. This paper challenges that view. We show, through theoretical analysis, simulation counterexamples, and empirical validation, that predictive models can remain operationally valid even when structurally incorrect. Our contributions are fourfold. First, we distinguish between directional agreement, ranking, and calibration, proving that sign alignment alone does not ensure efficiency when signals are mis-scaled. Second, we establish that structurally misspecified signals can still yield convex and viable efficient frontiers provided they maintain directional alignment with true returns. Third, we derive and empirically confirm a quantitative scaling law that shows how Sharpe ratios contract smoothly with declining alignment, thereby clarifying the role of calibration within the efficient set. Fourth, we validate these results on real financial data, demonstrating that predictive signals, despite structural imperfections, can support coherent frontiers. These findings refine the debate on causality in portfolio modeling. While causal inference remains valuable for interpretability and risk attribution, it is not a prerequisite for optimization efficiency. Ultimately, what matters is the directional fidelity and calibration of predictive signals in relation to their intended use in robust portfolio construction. ...

Order-Flow Filtration and Directional Association with Short-Horizon Returns ArXiv ID: 2507.22712 “View on arXiv” Authors: Aditya Nittur Anantha, Shashi Jain, Prithwish Maiti Abstract Electronic markets generate dense order flow with many transient orders, which degrade directional signals derived from the limit order book (LOB). We study whether simple structural filters on order lifetime, modification count, and modification timing sharpen the association between order book imbalance (OBI) and short-horizon returns in BankNifty index futures, where unfiltered OBI is already known to be a strong short-horizon directional indicator. The efficacy of each filter is evaluated using a three-step diagnostic ladder: contemporaneous correlations, linear association between discretised regimes, and Hawkes event-time excitation between OBI and return regimes. Our results indicate that filtration of the aggregate order flow produces only modest changes relative to the unfiltered benchmark. By contrast, when filters are applied on the parent orders of executed trades, the resulting OBI series exhibits systematically stronger directional association. Motivated by recent regulatory initiatives to curb noisy order flow, we treat the association between OBI and short-horizon returns as a policy-relevant diagnostic of market quality. We then compare unfiltered and filtered OBI series, using tick-by-tick data from the National Stock Exchange of India, to infer how structural filters on the order flow affect OBI-return dynamics in an emerging market setting. ...

Evaluating Transfer Learning Methods on Real-World Data Streams: A Case Study in Financial Fraud Detection ArXiv ID: 2508.02702 “View on arXiv” Authors: Ricardo Ribeiro Pereira, Jacopo Bono, Hugo Ferreira, Pedro Ribeiro, Carlos Soares, Pedro Bizarro Abstract When the available data for a target domain is limited, transfer learning (TL) methods can be used to develop models on related data-rich domains, before deploying them on the target domain. However, these TL methods are typically designed with specific, static assumptions on the amount of available labeled and unlabeled target data. This is in contrast with many real world applications, where the availability of data and corresponding labels varies over time. Since the evaluation of the TL methods is typically also performed under the same static data availability assumptions, this would lead to unrealistic expectations concerning their performance in real world settings. To support a more realistic evaluation and comparison of TL algorithms and models, we propose a data manipulation framework that (1) simulates varying data availability scenarios over time, (2) creates multiple domains through resampling of a given dataset and (3) introduces inter-domain variability by applying realistic domain transformations, e.g., creating a variety of potentially time-dependent covariate and concept shifts. These capabilities enable simulation of a large number of realistic variants of the experiments, in turn providing more information about the potential behavior of algorithms when deployed in dynamic settings. We demonstrate the usefulness of the proposed framework by performing a case study on a proprietary real-world suite of card payment datasets. Given the confidential nature of the case study, we also illustrate the use of the framework on the publicly available Bank Account Fraud (BAF) dataset. By providing a methodology for evaluating TL methods over time and in realistic data availability scenarios, our framework facilitates understanding of the behavior of models and algorithms. This leads to better decision making when deploying models for new domains in real-world environments. ...

Markowitz Variance May Vastly Undervalue or Overestimate Portfolio Variance and Risks ArXiv ID: 2507.21824 “View on arXiv” Authors: Victor Olkhov Abstract We consider the investor who doesn’t trade shares of his portfolio. The investor only observes the current trades made in the market with his securities to estimate the current return, variance, and risks of his unchanged portfolio. We show how the time series of consecutive trades made in the market with the securities of the portfolio can determine the time series that model the trades with the portfolio as with a single security. That establishes the equal description of the market-based variance of the securities and of the portfolio composed of these securities that account for the fluctuations of the volumes of the consecutive trades. We show that Markowitz’s (1952) variance describes only the approximation when all volumes of the consecutive trades with securities are assumed constant. The market-based variance depends on the coefficient of variation of fluctuations of volumes of trades. To emphasize this dependence and to estimate possible deviation from Markowitz variance, we derive the Taylor series of the market-based variance up to the 2nd term by the coefficient of variation, taking Markowitz variance as a zero approximation. We consider three limiting cases with low and high fluctuations of the portfolio returns, and with a zero covariance of trade values and volumes and show that the impact of the coefficient of variation of trade volume fluctuations can cause Markowitz’s assessment to highly undervalue or overestimate the market-based variance of the portfolio. Incorrect assessments of the variances of securities and of the portfolio cause wrong risk estimates, disturb optimal portfolio selection, and result in unexpected losses. The major investors, portfolio managers, and developers of macroeconomic models like BlackRock, JP Morgan, and the U.S. Fed should use market-based variance to adjust their predictions to the randomness of market trades. ...

Quantum generative modeling for financial time series with temporal correlations ArXiv ID: 2507.22035 “View on arXiv” Authors: David Dechant, Eliot Schwander, Lucas van Drooge, Charles Moussa, Diego Garlaschelli, Vedran Dunjko, Jordi Tura Abstract Quantum generative adversarial networks (QGANs) have been investigated as a method for generating synthetic data with the goal of augmenting training data sets for neural networks. This is especially relevant for financial time series, since we only ever observe one realization of the process, namely the historical evolution of the market, which is further limited by data availability and the age of the market. However, for classical generative adversarial networks it has been shown that generated data may (often) not exhibit desired properties (also called stylized facts), such as matching a certain distribution or showing specific temporal correlations. Here, we investigate whether quantum correlations in quantum inspired models of QGANs can help in the generation of financial time series. We train QGANs, composed of a quantum generator and a classical discriminator, and investigate two approaches for simulating the quantum generator: a full simulation of the quantum circuits, and an approximate simulation using tensor network methods. We tested how the choice of hyperparameters, such as the circuit depth and bond dimensions, influenced the quality of the generated time series. The QGAN that we trained generate synthetic financial time series that not only match the target distribution but also exhibit the desired temporal correlations, with the quality of each property depending on the hyperparameters and simulation method. ...

Determinants of Saving Behavior Among Employees in Dhaka, Bangladesh ArXiv ID: 2507.21254 “View on arXiv” Authors: Soumita Roy, Md Muntasir Kamal Dihan, Tasnimah Haque, Nafisa Nomani, Sadia Islam Preety Abstract Purpose With an emphasis on elements like financial knowledge, financial attitude, social influence, financial self-efficacy, and financial management practices, this study explores the factors that influence employees’ saving behavior in Dhaka, Bangladesh. We also welcome others to work on saving behavior, which is the main reason for publishing. The purpose is to make others aware of the methods for quantitative financial behavior analysis in Bangladesh. Design/methodology/approach The study uses a quantitative approach with a cross-sectional survey design. Data was collected from 40 participants through a structured questionnaire adapted from reliable sources. The questionnaire captured demographic information and used established items to measure the key variables. Data analysis included descriptive statistics, reliability analysis using Cronbachs alpha, and regression analysis to test the hypothesized relationships. Findings The results indicate that among the factors examined, only financial management practices had a significant positive relationship with saving behavior. Rest of the factors did not show significant relationships with saving behavior in this study sample. Limitation or Disclaimer It is still a work in progress, this paper is meant for pre-print with mostly incomplete and limited data. No data cleaning was performed, so it is very likely to include outliers and faulty data. Originality or value This study contributes to the limited research on saving behavior determinants in the Bangladeshi context, specifically among employees in the capital city of Dhaka. It explores the influence of multiple factors, including the rarely studied aspect of social influence. ...