Inquire UK 2025/10/05 to 2025/10/07 In-Person Autumn Seminar at Eastwell Manor, Ashford

This paper assesses the real-world expected returns of machine learning (ML) based trading strategies while accounting for transaction costs, post-publication decay, and modern liquidity conditions. While typical ML strategies lose about 57% of their performance when facing these real-world constraints, long short-term memory models remain profitable with net monthly returns of 1.39% and significant risk-adjusted alphas in real-world conditions. This paper is valuable for investment professionals because it provides realistic performance expectations for machine learning strategies, identifies which models can profit more with market frictions, and provides a framework on how the diffusion of ML technology gradually reduces profits.

Download The Expected Returns on Machine-Learning Strategies

Professor
Department of Business Studies and Economics Rheinland-Pfälzische Technische Universität (RPTU)

Vitor Azevedo is a professor at the Chair of Financial Management at RPTU Kaiserslautern-Landau. He earned his Ph.D. in finance, graduating summa cum laude from the Technical University of Munich (TUM). From 2018 to 2021, he continued at TUM as a postdoctoral researcher. His research focuses on empirical asset pricing, behavioral finance, sustainable finance, and quantitative finance, with a particular interest in applying artificial intelligence and machine learning techniques. He has published in leading journals such as the Journal of Banking and Finance and the Journal of Asset Management. Before his academic career, he worked in financial markets as a portfolio manager and stockbroker.

This talk explores the application of causal discovery algorithms to factor investing, addressing recent criticisms of correlation-based models. We create novel causal network representations of the S&P 500 universe and apply them to three investment scenarios. Our findings suggest that causal approaches can complement traditional methods in areas such as stock peer group identification, factor construction, and market timing. While causal networks offer new insights and sometimes outperform correlation-based methods in terms of risk-adjusted returns, they do not consistently surpass traditional approaches. The causal method though shows promise in identifying unique market relationships and potential hedging opportunities. However, its practical implementation presents challenges due to computational complexity and interpretation difficulties. Our study demonstrates the potential value of causal discovery in factor investing, while also identifying areas for further research and refinement.

Download Causal Network Representations in Factor Investing

Portfolio Engineer
Abu Dhabi Investment Authority

Clint Howard is a Portfolio Engineer at Abu Dhabi Investment Authority specializing in managing systematic global equity strategies. He has particular research experience in stock selection, portfolio construction, and machine learning. From 2022-2024, he served as a Senior Researcher in Robeco’s Quant Equity team in Rotterdam, where he developed new and enhanced existing systematic strategies for developed and emerging markets. From 2016-2022, he was a Researcher in Macquarie’s Systematic Investments Team in Sydney, where he focused on alpha research, infrastructure development, and assisting with portfolio management duties for Australian and Global equity funds.
Clint holds a PhD in Finance from the University of Technology Sydney. He completed a Bachelor of Engineering (Chemical & Biomolecular) and a Bachelor of Science (Mathematics) with First Class Honours and the University Medal at the University of Sydney. His work has been published in journals including the Financial Analysts Journal, Quantitative Finance, Journal of Portfolio Management, and Intelligent Systems in Accounting, Finance & Management.

Our paper explores the global financial impact of nonverbal communication by Federal Reserve Chairs during Federal Open Market Committee press conferences, focusing specifically on their facial expressions. Using machine learning to analyze video data, we find that positive facial sentiment significantly boosts international stock market returns, particularly in economies with higher financial development, non-English official language, or real-time access to press conferences. Investment professionals may find this research valuable as it highlights an overlooked dimension of central bank communication—nonverbal cues—and demonstrates how they can inform market expectations, reduce uncertainty, and enhance investment strategies in an interconnected global economy.

Xintong Zhan
Professor of Finance and Li-Dasan Endowed Chair
Fudan University School of Management (FDSM)

Xintong Zhan is currently a Professor of Finance and Li-Dasan Endowed Chair at Fudan University School of Management (FDSM), a fellow at Asian Bureau of Finance and Economics (ABFER), and a senior honorary fellow at Lau Chor Tak Institute of Global Economics and Finance at Chinese University of Hong Kong. She is both a CFA Charterholder and a CAIA Charterholder. Prof. Zhan’s research interests are in empirical asset pricing, sustainable finance, and derivatives. She currently serves as associate editors at China Accounting and Finance Review (CAFR) and International Review of Finance (IRF). She is also a board member of Financial Analyst Journal. Her work has appeared in a variety of academic journals including the Journal of Financial Economics, the Review of Financial Studies, the Journal of Financial and Quantitative Analysis, and Management Science. She is the Principal Investigator of several Hong Kong competitive RGC grants and many other research grants from both academic and industry sponsors such as The Canadian Derivatives Institute (CDI) and Geneva Institute for Wealth Management. She has received various research awards such as AAM–CAMRI Prize in Asset Management by Asia Asset Management and NUS, the ETF Research Academy Award by the Paris–Dauphine House of Finance and Lyxor Asset Management, Chicago Quantitative Alliance (CQA) Academic Competition Award, and the best paper awards at various academic conferences.

Jie (Jay) Cao
Professor of Finance
School of Accounting and Finance, The Hong Kong Polytechnic University

Professor Jie (Jay) Cao is currently a full professor of finance at the School of Accounting and Finance, Hong Kong Polytechnic University (PolyU). He also serves as an Advisory Council member for Monetary Research at the Hong Kong Institute for Monetary and Financial Research (HKIMR), an Academic and Accreditation Advisory Committee member for The Securities and Futures Commission (SFC) of Hong Kong, and an Academic Working Group member for the United Nations Sustainable Stock Exchanges (SSE) Initiative. He is an Associate Editor of Financial Management and Asia-Pacific Journal of Financial Studies, an Editorial Board Member of Financial Analysts Journal, a Co-Editor of The International Review of Finance, and an Editor of China Accounting and Finance Review. Before joining PolyU, he served as a tenured associate professor of finance at the Chinese University of Hong Kong (CUHK) Business School and worked there for 13 years.

Prof. Cao received his Ph.D. in Finance from the University of Texas at Austin in 2009 and BA in Economics from Peking University in 2002. His research interests center on empirical asset pricing, derivatives, and sustainable finance. His papers are published or forthcoming in top finance and management journals such as Journal of Financial Economics, Review of Financial Studies, Journal of Financial and Quantitative Analysis, and Management Science. He is the Principal Investigator of several Hong Kong competitive RGC grants and many other research grants from both academic and industry sponsors such as The Canadian Derivatives Institute (CDI) and Geneva Institute for Wealth Management. He has received various research awards such as AAM–CAMRI Prize in Asset Management by Asia Asset Management and NUS, the ETF Research Academy Award by the Paris–Dauphine House of Finance and Lyxor Asset Management, Chicago Quantitative Alliance (CQA) Academic Competition Award, and the best paper awards at several academic conferences such as the 28th Australian Finance & Banking Conference, 2020 FMA Consortium on Asset Management, 2020 Northern Finance Association Annual Conference, etc.

Mutual-fund performance is traditionally evaluated using alpha, which measures the utility gain of an unconstrained investor who has access to the fund in addition to the benchmark factors. We prove that the utility gain of shortsale-constrained investors is instead measured by achievable alpha, estimated using only those factors with strictly positive weight in the shortsale-constrained benchmark-factor portfolio. Empirically, active-fund management is less valuable for constrained investors: while 57.42 % of funds have positive traditional gross alpha, only 30.77 % have positive achievable gross alpha for a benchmark containing eight Vanguard funds. Achievable alphas predict fund flows, particularly during market turmoil.

Download Rethinking Mutual Fund Performance: From Traditional Alpha to Achievable Alpha

Professor
EDHEC

Raman Uppal is Professor of Finance at EDHEC Business School. He holds a bachelor’s degree in Economics (Honors) from St. Stephen’s College, Delhi University, and M.A., M.B.A., and Ph.D. degrees from The Wharton School of the University of Pennsylvania. Prior to working at EDHEC Business School, he was at London Business School and the University of British Columbia. He has held visiting positions at Catholic University (Leuven), the MIT Sloan School of Management, and the London School of Economics and Political Science, and has served as the Co-Director of the Financial Economics Program of the Centre for Economic Policy Research (CEPR).

His research focuses on optimal portfolio selection and asset allocation in dynamic environments, household and behavioral finance, valuation of securities in capital markets, risk management, and exchange rates. This research has been published in the American Economic Review, Journal of Finance, Review of Financial Studies, Journal of Economic Theory, Journal of Financial and Quantitative Analysis, Journal of International Money and Finance, and Management Science. He is currently an editor of the Journal of Systematic Investing and an associate editor of The Critical Finance Review. In the past, he has served as an editor of the Review of Financial Studies, Review of Finance, Review of Asset Pricing Studies, and as an associate editor of Management Science.

He has taught courses on Portfolio Choice and Asset Pricing, Capital Markets, International Financial Markets, Multinational Financial Management, Risk Management, and Corporate Finance to students in undergraduate, master’s, doctoral, and executive programs. He is the recipient of the Dean’s Advisory Board’s Outstanding Teaching Award for 1988 at The Wharton School, the Teaching Excellence Award for undergraduate teaching in 2000 at the Sauder School of Business at The University of British Columbia, the General Excellence Teaching Award for 2001/2002 at London Business School, the inaugural Excellence in Teaching Award in 2008 at London Business School, the Prize for Pedagogical Excellence at EDHEC Business School in 2015, 2017, and 2018, and numerous other prizes for his teaching and research.

There will be a social intermission.  Attendees will enjoy the opportunity to demonstrate both skill, strength and concentration through archery as well as the opportunity to learn about and enjoy falconry.

The paper investigates how monetary policy affects the size and persistence of rational stock market bubbles through the lenses of nonlinear dynamic models. Beyond its theoretical and empirical estimation contributions, the key takeaway for investment professionals consists of a regime-switching trading strategy that incorporates monetary policy and sentiment indicators—such as interest rates and abnormal trading volumes—to forecast market dynamics. The strategy performs strongly out-of-sample, systematically generating alpha for risk-averse investors, even after accounting for transaction costs. These results offer valuable insights for enhancing portfolio allocation and managing risk in a policy-sensitive investment environment.

Assistant Professor
IÉSEG School of Management and Baffi Carefin Bocconi University

Monia Magnani is an Assistant Professor of Finance at IÉSEG School of Management in Paris, where she teaches and investigates courses in fixed income and portfolio management, and a Research Fellow at Baffi Carefin, Bocconi University in Milan. She recently completed a postdoctoral fellowship at Bocconi University as part of the “Fin4Green – Finance for a Sustainable, Green and Resilient Society” project, funded by the Italian Ministry of University and Research.
In 2024, she earned a PhD in Finance from the University of Liverpool Management School, during which she was a Visiting Researcher at De Nederlandsche Bank (Dutch Central Bank). Prior to her doctoral studies, she obtained an MSc in Finance from Bocconi University and began her career in the Risk Management divisions of Italy’s two largest banking groups.
Her research—published in journals such as the Journal of Financial Research and the Journal of Asset Management—focuses on ESG integration in trading strategies, as well as the formation of rational bubbles and their implications for macroeconomic policy and asset prices.

The Next Accounting Frontier: Bringing Structure and Reliability to NLP
The study employs a unique database that uses NLP extractions of qualitative information about quantitative financial disclosures. It provides two use cases of the data on Global and US universes. The study highlights:

• Structuring NLP around accounting topics makes extracted text signals more interpretable and repeatable for investors.
• NLP signals show positive returns with low correlation to traditional factors, offering new sources of alpha.
• Investors can customize signals by selecting and weighing specific topics, enabling strategy-specific applications

Professor Emeritus
New York University

Joshua Livnat, PhD, CPA (inactive), is a Professor Emeritus of Accounting at the Stern School of Business, New York University, and was a Managing Director and Head of Research for Quantitative Management Associates (QMA, now PQS). His primary research areas have included capital markets, the effects of various accounting disclosure on stock prices, market anomalies and valuation issues. Prior to joining QMA, Joshua was on the faculty of NYU’s Leonard Stern School of Business, where he was Professor of Accounting and chairman of the department. Previously, he taught at Vanderbilt University, University of California at Berkeley, Northwestern University and Hebrew University in Jerusalem. Joshua co-authored the book Cash Flow and Security Analysis and has been published in many journals, including Journal of Accounting Research, Journal of Accounting & Economics, The Accounting Review, Journal of Finance, Journal of Portfolio Management and Financial Analysts Journal. He earned a BS in Mathematics and Statistics from Hebrew University and a PhD in Accounting from New York University.

Ian Paczek has been a generous supporter of Inquire UK for decades… Yet…. he has never before delivered a talk.  Ian has finally agreed to share some insights gleaned from his experience running back tests. 

““I’ve run 14,000,605 backtests…”
“… and how many succeed? Only one.”
Unlike Doctor Strange, we don’t have the benefit of an Infinity Stone in our research, so picking one perfect backtest is unjustified.

This is a talk about biases in portfolio management and quantitative research.
From cherry-picking backtest results to thesis drift and Bridget Jones investing, I plan to galumph my way through a landscape of different biases and on the way to suggest how one might combat them.

Senior Portfolio Manager
UBS Asset Management

Ian Pączek, PhD, CFA is a senior portfolio manager at UBS Asset Management, where he has worked since 2004. Prior to that he was a senior quantitative analyst at JP Morgan Fleming Asset Management. He was previously a member of the financial engineering and business analysis team at Lombard Risk Systems.

Ian is a member of the CFA Institute and the Advisory Panel of Inquire. Additionally, he is a CFA Charterholder and has earned the CFA Institute’s Certificate in ESG Investing.

Ian has an MA from the University of Cambridge, an MSc from Newcastle University and a PhD in statistical mechanics from Manchester University (UMIST). He was briefly a physicist at UWO in London, ON, Canada – where he learned to ski and skate – and at the Japan Atomic Energy Research Institute in Tokai Mura – where he learned judo and Japanese.

Ian has long been a dedicated supporter of Inquire UK and ran the program committee for many years. His photo is testament to newfound mastery of AI digital photo manipulation.

Measurement error in features affects the out-of-sample performance of portfolios and of return predictions. In a high-dimensional feature space, increasing model complexity under proper regularization can enhance the predictability of asset returns; however, the Sharpe ratio of a portfolio of assets and the R-squared of the prediction of the asset returns decrease monotonically and are convex as the noise level in features increases. Furthermore, when only a subset of features is observed, there is an optimal level of complexity beyond which incorporating additional features can degrade portfolio performance due to the effect of noise in the features. Thus, the marginal benefits of increasing model complexity will, at some point, start to diminish as the number of observed features increases. Our findings underscore a limited virtue of complexity in financial forecasting, where the performance of portfolios depends on the noise level in features, and where more complex models do not necessarily lead to better performance when features are not perfectly observed.

Download The Limited Virtue of Complexity in a Noisy World

PhD student
Oxford-Man Institute of Quantitative Finance, University of Oxford

Yuantao Shi is a PhD student at the Oxford-Man Institute of Quantitative Finance, University of Oxford. He received a Master’s degree in Statistics from University of Chicago in 2023, and Bachelor’s degree in Mathematics and Applied Mathematics from Fudan University in 2021. Yuantao’s research focuses on the intersection of statistics, finance, and machine learning.

This paper assesses the real-world expected returns of machine learning (ML) based trading strategies while accounting for transaction costs, post-publication decay, and modern liquidity conditions. While typical ML strategies lose about 57% of their performance when facing these real-world constraints, long short-term memory models remain profitable with net monthly returns of 1.39% and significant risk-adjusted alphas in real-world conditions. This paper is valuable for investment professionals because it provides realistic performance expectations for machine learning strategies, identifies which models can profit more with market frictions, and provides a framework on how the diffusion of ML technology gradually reduces profits.

Download The Expected Returns on Machine-Learning Strategies

Professor
Department of Business Studies and Economics Rheinland-Pfälzische Technische Universität (RPTU)

Vitor Azevedo is a professor at the Chair of Financial Management at RPTU Kaiserslautern-Landau. He earned his Ph.D. in finance, graduating summa cum laude from the Technical University of Munich (TUM). From 2018 to 2021, he continued at TUM as a postdoctoral researcher. His research focuses on empirical asset pricing, behavioral finance, sustainable finance, and quantitative finance, with a particular interest in applying artificial intelligence and machine learning techniques. He has published in leading journals such as the Journal of Banking and Finance and the Journal of Asset Management. Before his academic career, he worked in financial markets as a portfolio manager and stockbroker.

Drawing on neuroscience and cognitive psychology, we present a new view of momentum, highlighting how clear trends in prediction shape future expectations. We show a clear price trend significantly enhances the momentum strategy, yielding annualized returns of 12.68% in the highest trend clarity quintile, compared to 6.90% in the traditional momentum strategy. We demonstrate how trend clarity influences momentum trading intensity within the multi-period return chasing model of Hong and Stein (1999). A series of tests shows that chasing a clear price trend is an intrinsic aspect of momentum, distinct from existing explanations and present in the global markets.

Download  Trended Momentum 

Professor of Finance and Director of Research
University of Liverpool Management School

Charlie X. Cai is Professor of Finance and Director of Research at the University of Liverpool Management School. His research focuses on empirical asset pricing, behavioural finance, financial market anomalies, and the intersection of finance with psychology and data science. Charlie has published widely in leading finance and accounting journals and currently serves as a joint Editor-in-Chief of The British Accounting Review. He has held visiting appointments in Asia and regularly advises industry audiences on financial market behaviour.

This paper studies the performance of actively managed bond mutual funds across several different sectors of the bond market. We use a comprehensive sample of actively managed bond mutual funds from a dataset that is robust in the cross-section as well as the timeseries. Using this dataset, we examine the covariates that are associated with superior active bond fund performance. We construct several sets of variables to measure different aspects of managerial ability, including risk management, credit analysis, activeness, beta timing, liquidity provision, and family synergy. Given the large set of covariates, we employ machine learning methods including Boosted Regression Trees to select the best predictors of bond fund performance. Unlike equity funds, we find that risk management plays an important role in generating superior performance. In addition, funds that are better at credit analysis and charge lower fees outperform their peers. Our paper uniquely employs machine learning methods to study the covariates of successful active management of bond funds across several sectors of the bond market.

Paul J. Cinquegrana ’63 Endowed Chair in Finance, Director, Center for Financial Policy
University of Maryland

Russ Wermers is the Paul J. Cinquegrana ’63 Endowed Chair in Finance and Director of the Center for Financial Policy (CFP) at the University of Maryland at College Park. His research, published in leading scholarly journals, has developed new approaches to measuring and attributing the performance of mutual funds, pension funds, and private equity funds, which, among other applications, can be used to identify superior active funds. Professor Wermers consults for the asset management industry. He received his Ph.D. from the University of California, Los Angeles, in December 1995.