Quant Letter: August 2024, Week-4

Weekly (60th Edition)

I am working on a revolutionary (my opinion) alternative data and machine learning product called sov.ai. Investment managers can get full commercial access to 25+ ticker linked datasets, for 5% of the costs of equivalent industry datasets that you can find on AWS Market place and Snowflake.

If you're interested in a dataset that's not currently available, please let me know, and we can explore options to create it. You can reach me at d.snow@sov.ai.

The inspiration for this project came from observing opportunities for better integration among alternative data solutions for investment insights.

In other news, the top Reddit posts are back, I got around the API limits and it should be stable from now onwards.

ArXiv

Finance

Extreme DRL for Hedging: The article introduces EXtreme DRL (EX-DRL), a new method to improve the accuracy of extreme quantile predictions in Distributional Reinforcement Learning, improving financial risk management. (2024-08-22, shares: 7)

Asset Pricing Uncertainty: The study investigates no-arbitrage asset pricing under model uncertainty and short sales prohibitions, extending the single-period securities model to a multi-period one. (2024-08-23, shares: 4)

Causal Discovery in Financial Networks: The article presents a new method for building financial networks using quantile regression and a piecewise linear embedding scheme. This method uncovers intricate tail interactions in financial markets and identifies Bitcoin as the main influencer. (2024-08-22, shares: 3)

EURUSD Rate Forecast: The paper presents a new framework, IUS, that merges unstructured text and structured financial data to improve the accuracy of EUR/USD exchange rate forecasts. (2024-08-23, shares: 3)

Dynamic Real Estate Pricing: The study introduces an improved mathematical model for optimizing real estate pricing, taking into account variable demand, time value of money, and real estate value growth with development stage. (2024-08-22, shares: 2)

Controllable Market Generation: The article suggests a Diffusion Guided metaAgent (DiGA) model to enhance control and accuracy in generating order flow in financial markets, proving its usefulness for subsequent financial applications. (2024-08-23, shares: 4)

Commodity Pricing Optimization: The paper presents a new method using the hybrid neural network model CNN-BiGRU-SSA for precise prediction and optimization of cross-border commodity pricing strategies, demonstrating superior performance on various datasets. (2024-08-22, shares: 3)

Historical Trending

Deviations in Financial Trading: Autonomous trading bots using advanced algorithms can disrupt markets by deviating from traditional predictions, often favoring optimal solutions over equilibrium. (2024-08-21, shares: 6)

Cheyette Models for Interest Rate Options: Research suggests that using alternative volatility terms and models can improve the calibration of the 1Y caplet smile across different strike ranges, particularly a model with linear local volatility and uncorrelated variance. (2024-08-21, shares: 4)

SSRN

Recently Published

Quantitative

Geospatial Analytics in Real Estate: Geospatial analytics integration in real estate marketing enhances targeting and predictive modeling, despite data complexity and privacy issues. (2024-02-13, shares: 2.0)

Machine Learning for Yield Curve: A unique tree-growing algorithm can detect regime-shifting patterns in the yield curve, providing easy economic interpretation and computation, with macroeconomic variables predicting the yield curve when the short rate is high. (2024-07-01, shares: 2.0)

Machine Learning for FX Prediction: Research combining foreign exchange rate forecasting with machine learning finds that machine learning models can predict currency variations more accurately than traditional models. (2024-08-26, shares: 2.0)

AI Risk Management in Global Banking: The paper discusses the use of AI in risk management in global banking, offering real-world examples and suggestions for financial institutions to utilize AI for better risk management. (2024-08-27, shares: 2.0)

Recently Updated

Quantitative

Earnings Forecasting and Portfolios in the US: A study confirms the continued effectiveness of a 1993 financial engineering model, outperforming equity benchmarks from 2000-2022. (2024-08-18, shares: 3.0)

Global Market Portfolio: A study of a $150 trillion global market portfolio from 1970-2022 shows it is more stable than equities, despite a similar Sharpe ratio, with risks appearing larger in non-U.S. currencies. (2024-08-27, shares: 34.0)

Financial

Asset Pricing Models UK: The study compares the CAPM FamaFrench 5 factor and Hou et al. 2015 qfactor model in the UK, concluding that the qfactor model is more effective. (2024-07-01, shares: 5.0)

Predictive Ability Technical Trading Rules: The study analyzes the predictability of top equity indices using technical trading rules, determining that the effectiveness of these rules is not consistent and markets become unpredictable over time. (2023-08-12, shares: 3.0)

Predicting Price Movements: The study uses machine learning to identify trade traits for accurate prediction of market trends. (2024-06-01, shares: 2.0)

ArXiv ML

Recently Published

Data Quality Antipatterns: A study reveals that the sequence of cleaning data quality antipatterns significantly impacts the performance and interpretation of machine learning models in software analytics. (2024-08-22, shares: 12)

Amortized Bayesian Models: Research explores the use of neural network architectures in Bayesian Multilevel Models to enable efficient training and inference on unseen data sets, solving computational challenges and providing quick posterior inference. (2024-08-23, shares: 8)

AI Workflow Steering: The paper introduces Colmena, an AI system that adapts and optimizes computational workflows on supercomputers, improving performance in various scientific fields and encouraging the use of AI in scientific computing. (2024-08-26, shares: 7)

Historical Trending

Predictor-Corrector Method: The paper explores the theoretical aspects of classifier-free guidance (CFG), debunking misconceptions about its interaction with DDPM and DDIM, and explaining its function as a predictor-corrector method, offering a deeper understanding of CFG within the context of principled sampling methods. (2024-08-16, shares: 170)

Data Quality Metric for LLMs: The study introduces a measure called the diversity coefficient to formalize data quality in pre-training Large Language Models (LLMs), demonstrating its alignment with diversity and variability properties, and its usefulness in evaluating downstream model performance. (2023-06-24, shares: 50)

Entity-based Neural Topic Modeling: The study uses bimodal vector representations to improve entity-based neural topic modeling, resulting in better coherency metrics than existing models. (2023-01-06, shares: 42)

SST Multi-Scale Hybrid Forecasting: The paper introduces the State Space Transformer model for time series forecasting, which effectively captures global and local patterns, offering superior performance with less memory and computational cost. (2024-04-23, shares: 30)

Reference Policies in DPO: The research reveals that Direct Preference Optimization in large language models is sensitive to the KL divergence constraint and performs better with stronger reference policies. (2024-07-18, shares: 18)

RePec

Finance

Calendar Anomalies in Cryptocurrencies: The research shows significant changes in volatility and day-of-the-week effects on cryptocurrency returns, especially during the COVID-19 pandemic. (2024-08-28, shares: 15.0)

Enhancing deep learning accuracy with multicriteria optimization: The study introduces a machine learning model that uses multicriteria optimization to minimize data fitting errors across multiple datasets, enhancing stability and reducing bias. (2024-08-28, shares: 25.0)

Predicting US bank failures with machine learning: The research shows that simple machine learning methods like the KNN model, combined with PCA, can effectively predict bank failures, surpassing traditional statistical methods. (2024-08-28, shares: 20.0)

Price Indices in Commercial Real Estate: The article introduces a new method for creating property price indices using machine learning, which is more accurate but less stable with small samples. (2024-08-28, shares: 15.0)

Deep Learning for Newsvendor Problems: The study applies a deep learning algorithm to solve complex control models for issues related to dynamic replenishment, financial hedging, and competition in the context of newsvendor problems, showing effective risk reduction. (2024-08-28, shares: 12.0)

Historical Trending

Alpha Generation with Neuroscience: The research indicates that long-term exposure to high volatility can lead to underestimation of volatility, which can be exploited for stock return predictability. (2022-04-08, shares: 13.0)

Forecasting Long-Term Equity Premiums: The research concludes that cross-sectional global factor models provide better long-term country equity premium forecasts than time-series prediction models. (2022-03-04, shares: 10.0)

GitHub

Finance

Algorithmic Trading: Learn Algorithmic Trading Published by Packt is a guide on understanding and applying algorithmic trading strategies. (2019-12-13, shares: 788.0)

Quantitative Finance: Quantitative Finance book is a detailed guide on financial analysis and modeling using quantitative techniques. (2019-04-10, shares: 439.0)

PyTrendFollow: PyTrendFollow systematic futures trading using trend following teaches how to use Python for trend-following in futures trading. (2018-02-24, shares: 343.0)

Cloning Environment: A different way of cloning a conda environment offers a new approach to replicate a conda software environment. (2023-01-31, shares: 5.0)

CS Papers to Read: Papers from the computer science community to read and discuss is a collection of important computer science research papers for discussion and analysis. (2013-12-15, shares: 86260.0)

LinkedIn

Trending

Index Replication ADMM Algorithm: The author discusses their progress in Quantitative Investing Research, using the ADMM algorithm to reduce tracking error and enhance risk-adjusted returns. (2024-08-22, shares: 4.0)

NonPaywalled Portfolio Construction Articles: The author presents a collection of free articles on robust portfolio construction, covering topics from eigenanalysis to finance-related deep learning. (2024-08-22, shares: 10.0)

Special Issues on AI and Digital Finance: Management Science has introduced two special issues focusing on AI for Finance and Business Decisions and Digital Finance, to promote research in digital economics and FinTech. (2024-08-25, shares: 3.0)

Informative

Building Volatility Models: Thalesians Ltd and G-Research are organizing a seminar where Prof. Jack Jacquier will discuss building volatility models using equity and equity options data. (2024-08-25, shares: 3.0)

Hedge Fund Risk: The founder of Alpha Architect shares his experience of a failed hedge fund launch in 2008, which led to the establishment of his current business. (2024-08-22, shares: 4.0)

Misconceptions About HFT: The article dispels six common myths about high-frequency trading, highlighting its complexity and the necessity for technical skills and deep financial market knowledge. (2024-08-27, shares: 4.0)

Best Books for Algo Trading: The article suggests five books for individuals interested in learning about algorithmic trading and machine learning in finance. (2024-08-22, shares: 3.0)

Avoiding Pitfalls in Model Building: The article provides insights from assisting Quantitative Researchers in model building, pointing out common mistakes and ways to prevent them. (2024-08-22, shares: 3.0)

Podcasts

Quantitative

A Quant Auditor's Day: Siddhesh Acharekar is conducting an online session about the life of a Quantitative Auditor, targeting students and professionals interested in quant finance. (2024-08-27, shares: 12)

Zach Creighton: Quant Recruiter: Zach Creighton, founder of Hemans, shares his experience in recruiting for quantitative finance, dispels recruiter myths, and provides job hunting advice in the quant finance field. (2024-08-27, shares: 7)

ETF Innovation and Investing: Mike Green of Simplify Asset Management talks about ETF innovation and competition, the impact of passive investing on the economy, and the intricacies of debt, deficits, and modern fiat systems. (2024-08-25, shares: 6)

Twitter

Quantitative

Empirical Asset Pricing and Machine Learning: The article reviews the relationship between empirical asset pricing and machine learning in financial literature. (2024-08-25, shares: 5)

Backtesting Applications in Python: The article highlights the significance of backtesting and Python DIY solutions in creating strong trading strategies. (2024-08-23, shares: 4)

Systematic Hunt for Alpha: The ManGroup article introduces the systematic method of Alpha hunting in investment strategies. (2024-08-28, shares: 2)

Latest Research on Quant Investing: New research on quantitative investing, including alternative data and sentiment trading, has been published. (2024-08-27, shares: 2)

Deep Learning in Finance: A recently published survey paper explores the use of deep learning in finance. (2024-08-23, shares: 2)

Miscellaneous

Understanding Core Inflation: A recent handbook chapter discusses how core inflation is measured and why it's crucial for policymakers. (2024-08-26, shares: 0)

Carry in Favor: Campbell & Company's multiasset Carry strategy has shown to be an effective diversifier for traditional assets and other systematic strategies. (2024-08-23, shares: 0)

Causal Inference in Science: The article emphasizes the role of causal inference in enhancing the accuracy and reliability of scientific research. (2024-08-22, shares: 0)

Causal Features in Prediction: The article proposes that pinpointing causal features can improve the accuracy of return predictions compared to conventional feature selection techniques. (2024-08-22, shares: 0)

Reddit

Quantitative

Quant Researchers vs Data Scientists: The article weighs the pros and cons of using AI in healthcare. (2024-08-27, shares: 55.0)

Best Level 3 Data Provider: The piece investigates how climate change affects global food security and farming. (2024-08-28, shares: 47.0)

Nightmare Quant Developer Interview: The article analyzes the recent increase in cryptocurrency investments and its effects. (2024-08-22, shares: 307.0)

Buy the Dip Backtest Results: This article studies the impact of social media on public opinion and politics. (2024-08-24, shares: 531.0)

Similar Finance Jobs to Quant: The piece emphasizes the need for mental health awareness in schools and proposes enhancement methods. (2024-08-22, shares: 84.0)

Financial Data: The piece discusses the effects of climate change on worldwide food security, emphasizing the necessity for sustainable agriculture. (2024-08-27, shares: 94.0)

Interest Rate Valuation: The article investigates the growth of cryptocurrency, its impact on the global economy, and the associated risks. (2024-08-26, shares: 21.0)