Machine Learning-Based Systematic Investing in Agency Mortgage-Backed Securities

Date: September 19

With a total outstanding balance of more than $8 trillion, agency mortgage-backed securities (MBS) represent the second largest segment of the US bond market and the second most liquid fixed-income market after US Treasuries. Institutional investors have long participated in this market to take advantage of its attractive spread over US Treasuries, low credit risk, low transaction cost, and the ability to transact large quantities with ease. MBS are made of individual mortgages extended to US homeowners.

The ability for a homeowner to refinance at any point introduces complexity in prepayment analysis and investing in the MBS sector. Traditional prepayment modeling has been able to capture many of the relationships between prepayments and related factors such as the level of interest rates and the value of the embedded prepayment option, yet the manual nature of variable construction and sheer amount of available data make it difficult to capture the dynamics of extremely complex systems.

The long history and large amount of data available in MBS make it a prime candidate to leverage machine learning (ML) algorithms to better explain complex relationships between various macro- and microeconomic factors and MBS prepayments.

Moderated by Dr. Kathryn Wilkens, Nikhil Jagannathan and Leo Bao propose a systematic investment strategy using an ML-based mortgage prepayment model approach combined with a coupon allocation optimization model to create an optimal portfolio to capture alpha versus a benchmark.

Slide Deck

Leveraging Large Language Models (LLMs) in Finance

Date: August 17

Slide Deck

Revolutionizing the Financial Industry Through Python and Open Source

Date: August 15

Slide Deck

Unleashing the Power of Neural Networks: A Personal Journey into Creating and Harnessing a Neural Network for Trading Stocks

Date: July 20, 2023

In a world where artificial intelligence is becoming complex and gaining influence, creating and using a machine learning model is not only a technical endeavor but also a personal journey of exploration, challenges, and growth. Tom Pickel, Founder of Souppe will share his journey of building a neural network from the ground up.

In recent years, the field of artificial intelligence (AI) has witnessed unprecedented advancements, and neural networks have emerged as a transformative technology, as they are able to learn from data and make predictions without being explicitly programmed. Neural networks are used in a wide variety of applications, including image recognition, natural language processing, speech recognition and perhaps even investing.
We will provide a basic overview of the feed-forward neural network and discuss the basic steps involved in creating and using a neural network, such as data preparation, setting hyperparameters, creating the architecture, training, validating, testing, performing gridsearch and more. We will also discuss some of the challenges associated with neural networks and how to overcome them.

Tom will share his experience in creating a neural network using Python’s basic data science packages (Numpy and Pandas) for trying to predict movements in the stock market. We will mention loading data from stock markets, creating the Features and dataset, training networks, k-fold cross validation, choosing the best architectures, backtesting, choosing a trading strategy and more.

This webinar is intended for anyone interested in learning more about neural networks. No prior experience is required, and perhaps it will enable participants to embark on their own journey towards developing their own machine learning models.

Slide Deck

Large Language Models in Finance: Advances and Impact

Date: July 18, 2023

Alik Sokolov, CEO of Responsibli and Dr. Kathryn Wilkens, Founder of Pearl Quest will discuss the revolution of natural language processing in recent years, and how it applies to various areas of investment management. Our ability to work with unstructured text data, which is abundant in investment management, has undergone several evolutions from the late 2010's: from sequence-to-sequence models for machine translation, to the advent of transformers and transfer learning, to the recent breakthroughs achieved by Large Language Models popularized by Chat GPT. These changes will have a profound impact on investment management, and we will examine several case studies and applications. We will also examine the long-term future of various investment and wealth management roles, and especially the long-term impact on ESG and responsible investing.

Slide Deck

Can Large Language Models (ex. ChatGPT) Produce More Accurate Analyst Forecasts?

Date: June 21, 2023

Russ Goyenko, Associate Professor of Finance at McGill University discusses with Dr. Hossein Kazemi how large language models can, and soon they will produce more accurate analyst forecasts. Using textual information from a complete history of regular quarterly and annual (10-Q and 10-K) filings by U.S. corporations, we train machine learning algorithms and large language models, LLMs, to predict future earnings surprises. First, the length of MD&A section on its own is negatively associated with future earnings surprises and firm returns in the cross-section. Second, neither sentiment-based nor classic NLPs approaches are able to ``learn'' from the past managerial discussions to forecast future earnings. Third, only "finance-trained" LLMs have the capacity to "understand'' the contexts of previous discussions to predict both positive and negative earnings surprises, and future firm returns.

Our evidence indicates significant, and somewhat hidden in the complexity of presentations, informational content of publicly disclosed corporate filings, and superior (to human) abilities of more recent AI models to identify it.

We are proud to welcome Claude Perron, Founder,  FIAM as an Association Partner sponsoring today's event. FIAM's focus of its activities is on issues, challenges & opportunities resulting from the rapid deployment of disruptive technologies in Finance, particularly in the use of AI/ML in asset management. FIAM has chosen to promote the FDP designation as a pillar of its activities.

Due to proprietary nature of the presentation the slide deck is not provided.

The Best of Both Worlds: Forecasting US Equity Market Returns Using a Hybrid Machine Learning-Time Series Approach

Date: June 20, 2023

Predicting long-term equity market returns is of great importance for investors to strategically allocate their assets. Harshdeep Ahluwalia, Head of Asset Allocation, Americas for Vanguard Investment Strategy Group is one of the authors invited to discuss the exploration of machine learning methods to forecast US stock returns 10-years ahead and compare the results to the traditional Shiller regression-based forecasts more commonly used in the asset-management industry.

The authors found that Machine learning techniques can only modestly improve the forecast accuracy of a traditional Shiller cyclically adjusted price-to-earnings (CAPE) ratio model, and they actually result in worse performance than the vector autoregressive model (VAR)-based two-step approach introduced by Davis et al. (2018).

Harshdeep will discuss with Dr. Kathryn Wilkens, Curriculum Consultant for the FDP Institute, how the authors then implemented the VAR-based two-step approach of Davis et al. (2018) with machine learning techniques and allowed for unspecified nonlinear relationships (a hybrid ML-VAR approach). They found up to 56% improvement in real-time forecast accuracy for 10-year annualized US stock returns.

Slide Deck

Forking Paths in Empirical Studies

Date: May 23, 2023

Guillaume Coqueret, Associate Professor, Emlyon Business School and Dr. Hossein Kazemi, Senior Advisor, FDP Institute discuss the importance of small variations in the implementation protocol of applied studies. This presentation will share why we advocate the usefulness of reporting a wide range of outcomes in empirical work, based on many variations of design choices. This allows us to characterize the effects more exhaustively and leads to more robust conclusions. We illustrate these ideas in two studies: one on equity premium prediction and one on portfolio sorts (asset pricing anomalies, i.e., factors).

Slide Deck

How to Build Better Portfolios in Python Using Riskfolio-Lib

Date: April 18, 2023

Are you looking to optimize your investment portfolio to maximize returns and minimize risks? Do you want to learn how to use Python to build powerful investment strategies and manage your portfolio efficiently? Then, this webinar is for you!

In this webinar, we will introduce you to Riskfolio-Lib, a powerful open-source library for portfolio optimization in Python created by Dany Cajas. Riskfolio-Lib provides a range of portfolio optimization techniques, risk management tools, and performance analytics to help investors design and manage their portfolios effectively.

Dany Cajas will demonstrate some popular use cases and will discuss with FDP Charterholder, Cordell Tanny how this library has evolved. It includes not only traditional methods of optimization, but also contains some of the most modern techniques available.

Slide Deck

Deep Neural Net Applications on Trading and Risks

Date: April 13, 2023

Many cutting-edge and AI/ML technologies have matured from R&D into robust tools in the past few years. Successful applications tend to be correctly scoped to provide immediate impact (3-6 months) to resolve existing bottlenecks, and by re-imagining a fresh approach on ‘how things are done’, through re-working of critical but implicit/unquestioned assumptions.

Some examples include applications on information extraction and analytics for nowcasting, tracking and resolving data transformation and quality issues, untangling complex dependencies and internal structures on long-chains of processes, and machine learning to resolve massive calculation performance bottlenecks, to name but a few.

This talk between Dr. Hossein Kazemi, FDP Institute and Dr. Gary Wong, Artemis AG will focus on applying Deep Neural Net (DNN) on 2 areas:
• Using DNN to replicate computationally intensive models (complex models and/or long-dated trades) and boosting massive calculation performance (1M+ times speed up) for trading and risks, e-trading on structured products, and regulatory calculations such as CVA and FRTB – a game-changer to totally turnaround the cost/benefit analysis.
• Pattern recognition on markets – analyzing and nowcasting for volatility surfaces – both liquid and illiquid markets. Apart from applications in trading and risk management, generating realistic and arbitrage-free synthetic vol surfaces are useful for data benchmarking, estimating missing data points for illiquid markets, and generating realistic stress scenarios.

Slide Deck

How Should Long-Term Investors Form Portfolios?

Date: March 28, 2023

Nowadays, the availability of alternative data and their adoption in the asset management industry to improve investment decision processes is a widespread alternative approach. It comes with the help of Machine Learning and Artificial Intelligence techniques that allow the processing of these data, reducing their noise, and converging Big Data into Smart Data.

Traditional asset management techniques provide very little guidance to these questions. Academic studies are often rooted in 50 or 70-year-old portfolio theories, as are standard investment textbooks. Yet these theories have very little to do with real-life portfolio practice.

McGill University, Associate Professor of Finance, Russ Goyenko will discuss with Senior Advisor, FDP Institute, Dr. Hossein Kazemi, how long-term investors should form portfolios. They will delve into how they should evaluate securities, portfolios, and managers. Learn more about how they adapt to time-varying expected returns, volatilities, correlations, and many factors, signals, and strategies.

We are proud to have Claude Perron, Founder of FIAM as an Association Partner sponsoring today's event. FIAM's mission is to contribute to the development and competitiveness of the Quebec financial community and beyond.

Developing an Improved Investment Style Analysis Algorithm and My Journey to the FDP Designation by Cordell Tanny

Date: March 9, 2023

In this fireside chat and live demonstration, Cordell Tanny will share with Dr. Hossein Kazemi his journey through his financial services career and how it led him to the FDP designation. He will describe his experiences with the program and how the material learned in the curriculum helped him not just advance his knowledge of how machine learning and AI can be used in finance, but also how to apply many of the concepts to his daily workflow. By the end of the presentation, the audience will have a better understanding of the importance of the FDP designation and gain insights into the journey of obtaining the FDP Charter. The audience will also see how open-sourced python libraries and financial data from free sources can be used to build innovative quantitative models.

We are proud to have Claude Perron, Founder of FIAM as an Association Partner sponsoring today's event. FIAM's mission is to contribute to the development and competitiveness of the Quebec financial community and beyond.

Slide Deck

FDP Q2-2023 Candidate Orientation Session

Date: March 2, 2023

Slide Deck

From TradFi to DeFi Through the Lens of Experience

Date: February 23, 2023

The CAIA Boston Chapter is pleased to present a panel discussion of global leaders who have led significant financial service organizations through periods of change and innovation over the last couple of decades. Learn how they have taken, adapted, and augmented those skills in this new era of digital disruption and innovation 2.0, under the broad heading of DeFi. How do they assess the new opportunities within a sea of disruption, and learn why the underpinnings of risk management, process, and procedure remain so critically important for a sustainable business model.

CAIA Association CEO, Bill Kelly will explore this topic with Julia Bonafede, CFA, Co-Founder of Rosetta Analytics, Jane Buchan, CEO of Martlet Asset Management, Zoe Cruz , Founder & CEO of Menai Financial Group and Jennifer Murphy, Founder & CEO of Runa Digital Assets.  Thank you to Jennifer Tribush, CAIA Boston Chapter and Senior Vice President, Global Head of Alternatives Product, State Street for supporting this CAIA Association and FDP Institute educational webinar.

How is ESG Reshaping the Alternative Investment Business?

Date: February 15, 2023 

Today, ESG is no longer just an acronym but a reality, expected to reach a third of assets under management by 2025. It is spreading across all asset classes, including the less liquid ones. ESG now defines the new frontier of alternative investments. Our speakers will be discussing the recent trends in alternative investments, how ESG impacts the alternative investment business and the important role of data.

Laura Merlini, Managing Director, EMEA will moderate a discussion with Mehrzad Mahdavi, Executive Director of FDP Institute along with Florence Angles, Managing Principal, Capco and Julia Bonafede, Co-Founder, Rosetta Analytics.