Quant Letter: July 2023, Week 2

Weekly quantitative finance newsletter

This week we scanned through 8,564 links to produce the 140 links you can find below. Please give the a like ♡ on the app if you still enjoy the content, or a comment if you wish for something to change.

I have finally obtained access to the API version of GPT-4, so the results looks better than last week.

Informed Trading Intensity

The leading article this week is “Informed Trading Intensity” forthcoming in the Journal of Finance. What I primarily like is the use of ML to construct indicators. ML is great at developing features/factors for downstream research and it’s also its most frequent use case in asset management.

SSRN

Recently Updated

Quantitative

Measuring Informed Trading Intensity with ML: A machine learning technique is used to create a new measure of informed trading intensity, which rises before significant announcements and affects return reversal and asset pricing. (2021-06-15, shares: 2.0)

Self-Supervised Learning for Diversifying Portfolios: The article discusses the use of self-supervised learning to analyze financial time series data for portfolio diversification, specifically for index tracking and minimum variance portfolio optimization. (2022-08-15, shares: 2.0)

SMARTboost: Efficient Tabular Learning: SMARTboost, a new machine learning model, is designed to fit complex functions in large dimensions, adjust model complexity, manage various features, and cater to specific financial needs. (2021-12-06, shares: 398.0)

ML Predicts Fund Performance: Machine learning can forecast top-performing mutual funds using fund characteristics, particularly fund momentum and fund flow. (2021-12-07, shares: 5001.0)

Unsupervised ML in Financial Time-Series Analysis: The study merges ontological methodology and temporal clustering to detect structural changes and crucial periods in financial time series, building on prior research in commodity markets. (2023-05-14, shares: 2.0)

Forecasting Financial Risk with Quantile RF: The study introduces a financial risk forecasting model using Generalized Quantile Random Forests, which offers competitive risk and shortfall forecasts and generates appealing Sharpe, Sortino, and Omega ratios. (2023-01-17, shares: 2.0)

Estimating Panel Data Models with Heteroskedasticity: The research provides a condition for accurately estimating structural parameters in panel data models with cross-sectionally heteroskedastic data. (2023-01-15, shares: 3.0)

Financial

Side-by-Side Management and Bond Fund Performance: The research reveals that bond mutual funds managed by managers with performance-based fees receive fewer fund flows and inflate their asset values, indicating potential conflicts of interest in side-by-side management. (2023-07-03, shares: 5.0)

Time-Varying Equity Premia & Sentiment: From 1990 to 2022, equity market returns can be predicted using a simple model, with higher returns following high implied volatility and lower returns after high market sentiment. (2023-06-19, shares: 108.0)

Thematic Investing: Fund Performance: Mutual fund managers can outperform by using thematic investment strategies, with a higher thematic concentration index leading to significant superior performance. (2022-08-05, shares: 427.0)

Market Concentration & Wealth Dynamics: A new theory suggests that financial market concentration is dynamic, with risk and wealth distribution determining market power, and wealth changing over time due to strategic portfolio decisions. (2021-11-28, shares: 153.0)

International Corporate Bond Returns Prediction with ML: Machine learning is used to forecast global corporate bond returns, showing varying influential factors in U.S. and non-U.S. markets and different levels of bond integration among countries. (2022-06-27, shares: 239.0)

Credit Market Fragility: Evidence from Asset Demand System: A two-layer asset demand framework is created to study the fragility of the corporate bond market, using microdata to assess the impact of unconventional monetary and liquidity policies on asset prices and institutions. (2022-12-12, shares: 372.0)

ML in Financial Markets: A Survey: A review of the emerging literature on machine learning in financial markets identifies promising research areas and provides insights for financial economists and machine learners. (2023-07-01, shares: 14.0)

Recently Published