Topic

Speaker

10:00 - 10:25

Registration

Registration

10:25 - 10:30

Opening Remarks

Farouk Hadeed, President, CUATS

10:30 - 11:15

KEYNOTE 1

Equity Order Flow and Prices

How does equity order flow affect prices? Do all type of traders have the same impact on prices? We use a unique set of data with explicit information on the trader type and side of order coupled with linear and non-linear techniques, to shed light on these empirical questions. Our analysis covers different geographies, data at the security level and at the portfolio level (i.e. investment styles/factors, industries/sectors), as well as different trading frequencies. We find that different types of traders have distinct impact on prices. We also find that linear models are helpful in distilling useful information, but machine learning techniques help us identify intuitive patterns in the data that are not attainable otherwise. Finally, we explore how information of order flow can be used in the context of systematic trading and find supportive evidence across different actionable settings.    

Daniel Giamouridis, PhD, Managing Director, Global Head of the Systematic Strategies Group, BofA Securities Global Equities

Daniel Giamouridis, PhD, Managing Director, is Global Head of the Systematic Strategies Group in BofA Securities Global Equities. Dr. Giamouridis’ career has focused on Quantitative investment decision making and analysis, in particular the design and implementation of systematic investment strategies. Prior to joining BofA Securities, Dr. Giamouridis was an Associate Professor of Finance at AUEB and the Academic Consultant of the Global Equity Quantitative Research Group at Citi. His investment experience and published research span systematic alpha, risk premia strategies and factor investing, pension asset management, investment risk measurement/management, and quantitative portfolio execution/trading. Dr. Giamouridis holds a PhD in finance from Bayes Business School-City, University of London and an MEng in naval architecture and marine engineering from NTUA. He is an affiliate of Bayes Business School-City, University of London and a co-editor of the Financial Analysts Journal.

11:15 - 12:00

KEYNOTE 2

How Universality Shapes Market Microstructure

How do trades move prices? This question lies at the heart of research in market microstructure, and kept researchers busy for several decades. I will provide an alternative angle to discuss this problem, building on the idea of universality. I will show in particular how many results from the last 40 years of research can be derived within a compact, principled framework based on a small number of underlying assumptions. The framework provides indications on how to build an impact model that consistently encompasses both temporal and cross-sectional aspects of the price response, an open problem of paramount importance in the field.

Iacopo Mastromatteo, Head of Directional Portfolio Construction, Capital Fund Management

Iacopo Mastromatteo is Executive Director at Capital Fund Management where he is in charge of Transaction Costs Analysis, design of execution models and portfolio construction. He holds a PhD in Statistical Physics from the International School for Advanced Studies of Trieste. His main interests involve statistical learning and market microstructure. He has contributed to the research in these fields with more than twenty research papers. 

12:00 - 13:00

Lunch in Sidney Sussex College Hall

Lunch in Sidney Sussex College Hall

13:00 - 13:45

KEYNOTE 3

Bond Portfolio Construction : a Mixed-Integer Optimisation Problem

The corporate bond indices, built by market index providers to serve as investment benchmarks, contain a great many securities, and are for that reason difficult to replicate. The art is to construct an investible portfolio that captures the general price trend among the several thousands of securities in the index, being limited to selecting few of them. This paper describes a practical approach to this, which combines a well-established portfolio construction technique known as stratified sampling with a modern bond risk measure named the Duration Times Spread.
The key idea is to divide the index members into samples related to distinct sources of risk that play in the corporate bond markets, and build small subsamples that capture those risks.
As the Duration Times Spread conveys linear- as well as non-linear bond price behaviour, it proves an effective measure in the portfolio building process.

Marielle de Jong, Associate Professor, Grenoble Ecole de Management

Marielle is Associate Professor at the Grenoble Ecole de Management since October 2020, and the academic director of the DBA USA school at GEM. She lectures in the fields of portfolio management, fixed income and sustainable investing.

Marielle has worked as a researcher in the investment management industry for 25 years. She started her career in the City of London in 1994, and moved to Paris in 1997, joining (HSBC) Sinopia, and in 2011 Amundi, where she headed the fixed-income quant research team.

She carries out research projects in areas such as bond portfolio construction, sustainable investing and liquidity scoring, and publishes her work in international journals. She is the editor-in-chief of the Journal of Asset Management.

She holds an MSc in econometrics from the Erasmus University of Rotterdam, an MSc in operational research from Cambridge University (UK), and a PhD in finance from the University of Aix-Marseille. She has defended her HDR (Habilitation à Diriger des Recherches) in 2022.

13:45 - 14:30

KEYNOTE 4

Building AI Models for Finance: Guiding Machines to Search for Solutions

I provide an overview of how recent development in AI can be utilized to answer core questions in finance. I then touch on deep reinforcement learning (RL) for portfolio management before focusing on panel trees for clustering assets and constructing pricing kernels. 

Specifically, in Cong, Tang, Wang, and Zhang (2019), we directly optimize the objectives of portfolio management via deep reinforcement learning. We develop multi-sequence, attention-based neural-network models tailored for the distinguishing features of financial big data, while allowing interactions with the market states and training without labels. Such AlphaPortfolio models yield stellar out-of-sample performances (e.g., Sharpe ratio above two and over 13% risk-adjusted alpha with monthly re-balancing) that are robust under various market conditions and economic restrictions (e.g., exclusion of small and illiquid stocks). We further demonstrate AlphaPortfolio's flexibility to incorporate transaction costs, state interactions, and alternative objectives, before applying polynomial-feature-sensitivity analysis to uncover key drivers of investment performance, including their rotation and nonlinearity. Overall, we highlight the utility of deep reinforcement learning in finance and "economic distillation" for model interpretation.

In a separate study, Cong, Feng, He, and He (2022), we develop a new class of tree-based models (P-Tree) for analyzing (unbalanced) panel data utilizing global (instead of local) split criteria that incorporate economic guidance to guard against overfitting while preserving interpretability. We grow a P-Tree top-down to split the cross section of asset returns to construct stochastic discount factors and test assets, generalizing sequential security sorting and visualizing (asymmetric) nonlinear interactions among firm characteristics and macroeconomic states. Data-driven P-Tree models reveal that idiosyncratic volatility and earnings-to-price ratio interact to drive cross-sectional return variations in U.S. equities; market volatility and inflation constitute the most critical regime-switching that asymmetrically interacts with characteristics. P-Trees outperform most known observable and latent factor models in pricing individual stocks and test portfolios, while delivering transparent trading strategies and risk-adjusted investment outcomes (e.g., out-of-sample annualized Sharp ratios of about 3 and monthly alpha around 0.8%).

Lin William Cong, Professor of Finance, Cornell University

Lin William Cong is the Rudd Family Professor of Management (endowed faculty chair by the Rudd Family Foundation) and Tenured Professor of Finance at the Johnson Graduate School of Management at Cornell University SC Johnson College of Business. He is also the founding faculty director for the FinTech Initiative at Cornell and a research associate at the National Bureau of Economic Research. 

Cong researches on financial economics, information economics, FinTech and Economic Data Science, Entrepreneurship, and China. His academic interests include financial innovation, mechanism and information design, blockchains, cryptocurrencies, digital economy, real options, financial policy and markets in China, machine learning, AI, and alternative data. His recent work has focused on the intersection of technology, data science, and finance. 

Cong earned a Ph.D. in Finance and a MS in Statistics from Stanford University, where he served as the president of Ph.D. students association, received the Asian American Award for Graduate Leadership, and was recognized with the Lieberman Fellowship for outstanding contributions in research, teaching, and university service. He also holds dual degrees from Harvard University where he graduated summa cum laude and top in the Physics department, with an A.M. in Physics, an A.B. in Math & Physics, a minor in Economics, and a language citation in French.

14:30 - 15:00

Break with Refreshments

Break with Refreshments

15:00 - 15:45

KEYNOTE 5

Sentiment Analytics in Asset Management

In this presentation, Matthias will talk about real-world applications of quant models inside a discretionary driven investment process in asset management. He will focus on how to build models based on NLP-generated sentiment from news articles for predicting multi-asset price movements, also mentioning some of the difficulties quants can face.

Matthias Uhl, Executive Director and Head of Analytics and Quant Modelling, UBS Asset Management

Matthias is the Head of Analytics & Quantitative Modelling (AQM) in Investment Solutions at UBS Asset Management, the global lead for sentiment analytics at UBS Group, and a lecturer at the University of Zurich. Previously, he was Chief Investment Officer at FLYNT Bank AG, has worked as quantitative strategist in the CIO Office at UBS Wealth Management, as FX and rates trader at UBS Investment Bank, as commercial banker at Deutsche Bank, and as economist at KOF Swiss Economic Institute.

Matthias holds a Ph.D. in applied macroeconomics and behavioral finance from ETH Zurich, a Master of Science from Oxford University and two Bachelor of Arts degrees from the American University of Paris. Matthias has published his research in various academic journals, such as the Journal of Portfolio Management, Finance Research Letters, Journal of Derivatives, and Journal of Behavioral Finance, among others.

15:45 - 16:30

KEYNOTE 6

The Challenge of Small Data Problems in Finance

Over the last decade we have seen significant advances in machine learning across a wide range of fields. In many cases, this has come from applying very complex models to extremely large datasets often containing millions of examples. These applications are often described as ‘big data’ problems. However, there is a related category of problems where the amount of available data to train a machine learning model is fundamentally limited. "Small data" problems are very common in finance and need to be approached in a very specific way since in most cases, techniques designed to solve big data problems simply do not work well when applied to small data sets. In this talk, Chris will discuss some examples of small data problems in finance and outline some of the approaches that can be applied to address the challenges posed by small data.

Dr. Chris Longworth, Investment Director, GAM Systematic

Dr Chris Longworth is an Investment Director at GAM and heads the investment team at GAM Systematic Cambridge. Chris has over a decade of experience in systematic investment and leads the research and development of the team's Core Macro programme.  He is also the author of the longest standing model traded within that programme. In addition, Chris has led the effort to incorporate novel or alternative data sources into the portfolio as well as the expansion of the team’s tradable universe into new and unconventional markets. Prior to joining GAM Systematic, Chris was part of the Machine Intelligence Laboratory at Cambridge University, where he obtained MPhil and PhD degrees. He also holds a BSc in Computer Science from Royal Holloway, University of London.

16:30 - 16:35

Closing Remarks

Farouk Hadeed, President, CUATS

16:45 - 18:30

Networking Reception

Networking Reception