量化前沿速递：机器学习[20250914]

机器翻译，仅供参考！可使用微信自带翻译功能自行翻译

更多文献获取请关注公众号：量化前沿速递

文献汇总

[1] MM DREX

MM DREX

来源:ARXIV_20250908

[2] Volatility Modeling via EWMA Driven Time Dependent Hurst Parameters

基于EWMA驱动的时变赫斯特参数的波动率建模

来源:ARXIV_20250909

[3] Predicting Market Troughs

预测市场困境

来源:ARXIV_20250909

[4] Nested Optimal Transport Distances

嵌套最佳运输距离

来源:ARXIV_20250909

[5] Reinforcement Learning for Arbitrage Strategies in Stock Index Futures

股指期货套利策略的强化学习

来源:SSRN_20250909

[6] Automated Trading System for Straddle Option Based on Deep Q Learning

基于深度Q学习的跨期期权自动交易系统

来源:ARXIV_20250911

[7] FinZero

FinZero 的

来源:ARXIV_20250911

[8] Machine Learning with Multitype Protected Attributes

具有多类型保护属性的机器学习

来源:ARXIV_20250911

[9] An Interpretable Deep Learning Model for General Insurance Pricing

通用保险定价的可解释深度学习模型

来源:ARXIV_20250911

[10] Causal PDE Control Models

因果PDE控制模型

来源:ARXIV_20250912

[11] Application of Natural Language Processing in Unstructured Financial Data: A Comprehensive Survey and Implementation Framework

自然语言处理在非结构化金融数据中的应用：一个全面的调查和实现框架

来源:SSRN_20250912

[12] Machine Learning for Daily Return Direction Forecasting: A Comparative Study with Explainable AI Insights

每日回报方向预测的机器学习：与可解释人工智能见解的比较研究

来源:SSRN_20250912

[13] Optimal Ridge Regressions are Singular Value Decomposition

最优岭回归是奇异值分解

来源:SSRN_20250912

[1] MM DREX

标题:MM DREX

作者:Yang Chen, Yueheng Jiang, Zhaozhao Ma, Yuchen Cao Jacky Keung, Kun Kuang, Leilei Gan, Yiquan Wu, Fei Wu

来源:ARXIV_20250908

Abstract : The inherent non stationarity of financial markets and the complexity of multi modal information pose significant challenges to existing quantitative trading models. Traditional methods relying on fixed structures and unimodal data struggle to adapt to market regime shifts, while large language model (LLM) driven solutions   despite their multi modal comprehension   suffer from static strategies and homogeneous expert......(摘要翻译及全文见知识星球)

Keywords : 

[2] Volatility Modeling via EWMA Driven Time Dependent Hurst Parameters

标题:基于EWMA驱动的时变赫斯特参数的波动率建模

作者:Jayanth Athipatla

来源:ARXIV_20250909

Abstract : We introduce a novel rough Bergomi (rBergomi) model featuring a variance driven exponentially weighted moving average (EWMA) time dependent Hurst parameter  H t , fundamentally distinct from recent machine learning and wavelet based approaches in the literature. Our framework pioneers a unified rough differential equation (RDE) formulation grounded in rough path theory, where the Hurst parameter dynamically adapts to evolving......(摘要翻译及全文见知识星球)

Keywords : 

[3] Predicting Market Troughs

标题:预测市场困境

作者:Peilin Rao, Randall R. Rojas

来源:ARXIV_20250909

Abstract : This paper provides robust, new evidence on the causal drivers of market troughs. We demonstrate that conclusions about these triggers are critically sensitive to model specification, moving beyond restrictive linear models with a flexible DML average partial effect causal machine learning framework. Our robust estimates identify the volatility of options implied risk appetite and market liquidity as key causal drivers, relationships......(摘要翻译及全文见知识星球)

Keywords : 

[4] Nested Optimal Transport Distances

标题:嵌套最佳运输距离

作者:Ruben Bontorno, Songyan Hou

来源:ARXIV_20250909

Abstract : Simulating realistic financial time series is essential for stress testing, scenario generation, and decision making under uncertainty. Despite advances in deep generative models, there is no consensus metric for their evaluation. We focus on generative AI for financial time series in decision making applications and employ the nested optimal transport distance, a time causal variant of optimal transport distance, which is......(摘要翻译及全文见知识星球)

Keywords : 

[5] Reinforcement Learning for Arbitrage Strategies in Stock Index Futures

标题:股指期货套利策略的强化学习

作者:Min Dai,Yuchao Dong,Linfeng Li

来源:SSRN_20250909

Abstract : We propose a reinforcement learning approach to find arbitrage strategies in stock index futures, formulated as an optimal switching problem. We adopt an exploratory stochastic control framework with an entropy-regularized reward function and design an interpretable learning algorithm. A policy improvement theorem is established. Both simulation experiments and empirical analysis validate the effectiveness of our algorithm. Furthermore, our approach is applicable......(摘要翻译及全文见知识星球)

Keywords : 

[6] Automated Trading System for Straddle Option Based on Deep Q Learning

标题:基于深度Q学习的跨期期权自动交易系统

作者:Yiran Wan, Xinyu Ying, Shengzhen Xu

来源:ARXIV_20250911

Abstract : Straddle Option is a financial trading tool that explores volatility premiums in high volatility markets without predicting price direction. Although deep reinforcement learning has emerged as a powerful approach to trading automation in financial markets, existing work mostly focused on predicting price trends and making trading decisions by combining multi dimensional datasets like blogs and videos, which led to high computational......(摘要翻译及全文见知识星球)

Keywords : 

[7] FinZero

标题:FinZero 的

作者:Yanlong Wang, Jian Xu, Fei Ma, Hongkang Zhang, Hang Yu, Tiantian Gao, Yu Wang, Haochen You, Shao-Lun Huang, Danny Dongning Sun, Xiao-Ping Zhang

来源:ARXIV_20250911

Abstract : Financial time series forecasting is both highly significant and challenging. Previous approaches typically standardized time series data before feeding it into forecasting models, but this encoding process inherently leads to a loss of important information. Moreover, past time series models generally require fixed numbers of variables or lookback window lengths, which further limits the scalability of time series forecasting. Besides, the......(摘要翻译及全文见知识星球)

Keywords : 

[8] Machine Learning with Multitype Protected Attributes

标题:具有多类型保护属性的机器学习

作者:Ho Ming Lee, Katrien Antonio, Benjamin Avanzi, Lorenzo Marchi, Rui Zhou

来源:ARXIV_20250911

Abstract : Ensuring equitable treatment (fairness) across protected attributes (such as gender or ethnicity) is a critical issue in machine learning. Most existing literature focuses on binary classification, but achieving fairness in regression tasks such as insurance pricing or hiring score assessments is equally important. Moreover, anti discrimination laws also apply to continuous attributes, such as age, for which many existing methods are......(摘要翻译及全文见知识星球)

Keywords : 

[9] An Interpretable Deep Learning Model for General Insurance Pricing

标题:通用保险定价的可解释深度学习模型

作者:Patrick J. Laub, Tu Pho, Bernard Wong

来源:ARXIV_20250911

Abstract : This paper introduces the Actuarial Neural Additive Model, an inherently interpretable deep learning model for general insurance pricing that offers fully transparent and interpretable results while retaining the strong predictive power of neural networks. This model assigns a dedicated neural network (or subnetwork) to each individual covariate and pairwise interaction term to independently learn its impact on the modeled output while......(摘要翻译及全文见知识星球)

Keywords : 

[10] Causal PDE Control Models

标题:因果PDE控制模型

作者:Alejandro Rodriguez Dominguez

来源:ARXIV_20250912

Abstract : Classical portfolio models collapse under structural breaks, while modern machine learning allocators adapt flexibly but often at the cost of transparency and interpretability. This paper introduces Causal PDE Control Models (CPCMs), a unifying framework that integrates causal inference, nonlinear filtering, and forward backward partial differential equations for dynamic portfolio optimization. The framework delivers three theoretical advances  (i) the existence of......(摘要翻译及全文见知识星球)

Keywords : 

[11] Application of Natural Language Processing in Unstructured Financial Data: A Comprehensive Survey and Implementation Framework

标题:自然语言处理在非结构化金融数据中的应用：一个全面的调查和实现框架

作者:Matthew Anyiam

来源:SSRN_20250912

Abstract : The financial sector generates approximately 2.5 quintillion bytes of data daily, with 80-90% being unstructured text (FactSet, 2021). This research presents a comprehensive analysis of Natural Language Processing (NLP) applications in processing unstructured financial data, demonstrating how modern techniques including transformer-based models (BERT, GPT) achieve 15-20% accuracy improvements over traditional methods. We analyze implementations across sentiment analysis, risk detection, and automated......(摘要翻译及全文见知识星球)

Keywords : Natural Language Processing, Financial Analysis, Unstructured Data, FinBERT, Sentiment Analysis, Large Language Models

[12] Machine Learning for Daily Return Direction Forecasting: A Comparative Study with Explainable AI Insights

标题:每日回报方向预测的机器学习：与可解释人工智能见解的比较研究

作者:Krzysztof Płachta

来源:SSRN_20250912

Abstract : This study addresses two interrelated objectives. First, we conduct a comparative evaluation of six supervised learning models: Lasso, Random Forest, LightGBM, LSTM, and two feedforward neural networks, for forecasting the next-day direction of SPY returns. Using a long-horizon, expanding-window backtesting framework, we generate strictly out-of-sample forecasts and assess model performance based on risk-adjusted metrics, with the Sortino ratio as the primary......(摘要翻译及全文见知识星球)

Keywords : machine learning, financial forecasting, algorithmic investment strategies, testing architecture, random forest, neural networks, LSTM, LightGBM, Lasso, explainable AI, feature importance JEL codes: C45, C53, G11, G17

[13] Optimal Ridge Regressions are Singular Value Decomposition

标题:最优岭回归是奇异值分解

作者:Irene Aldridge

来源:SSRN_20250912

Abstract : Researchers often need to identify which variables matter most when analyzing data, a challenge known as feature selection. Ridge regression has been widely used to select the most important features. Meanwhile, eigenvalue-based methods have become increasingly popular for uncovering the underlying simplified structure hidden within complex datasets. This paper shows that eigenvalue methods are, in fact, optimal ridge regression models. Eigenvalue......(摘要翻译及全文见知识星球)

Keywords :