A collection of our white papers and peer-reviewed research articles related to macro / multi-asset investor behavior, hedging, risk regimes, liquidity risk, private assets, portfolio construction, and more.
By Megan Czasonis, Huili Song, and David Turkington
We show that LLMs can effectively extrapolate from disparate domains of knowledge to reason through economic relationships, and that this may have advantages over narrower statistical models.
Fundamentally, large language models (LLMs) and numerical models both learn patterns in training data. However, while traditional models rely on narrowly curated datasets, LLMs can extrapolate patterns across disparate domains of knowledge. In new research, we explore whether this ability is valuable for predicting economic outcomes. First, we ask LLMs to infer economic growth based on hypothetical conditions of other economic variables. We then use our Model Fingerprint framework to interpret how they use linear, nonlinear, and conditional logic to understand economic linkages. We find that their reasoning is intuitive, and it differs meaningfully from the reasoning of statistical models. We also compare the accuracy of the models’ reasoning using historical data and find that the LLMs infer growth outcomes more reliably than the statistical models. These results suggest that LLMs can effectively reason through economic relationships and that cross-domain extrapolation may add value above explicit statistical analysis.
By William Kinlaw, Mark Kritzman, and David Turkington
Conventional statistics hide important realities that investors need to know.
The correlation coefficient often fails to capture what really matters to investors. There are two reasons for this. First, investors often measure correlations using monthly data and assume that they also hold over one-year, five-year or ten-year periods. Unfortunately, in the real world, they often don’t. The second reason has to do with a fundamental misconception about diversification. The fact is, investors don’t always want it. Sure, they want it on the downside, in order to offset the poor performance of one or more assets. But on the upside they prefer all assets to rise in unison, which is the opposite of diversification. Put differently, they’d be happy to place their eggs, conveniently, in a single basket provided nobody steals it. Our research shows that correlations can vary through time based on a range of conditions including the level of interest rates, the degree of turbulence in financial markets, and the performance of major equity markets. Overall, our findings challenge the notion that returns evolve as a simple “random walk,” a critical pre-condition without which we must interpret the correlation coefficient distrustfully. To address these issues, we introduce the notion of co-occurrence and offer a new perspective on how investors should diversify portfolios.
By Megan Czasonis, Mark Kritzman, and David Turkington
We propose a new currency hedging technique called full-scale hedging, which accounts for the complexities of diversification.
Diversification is nuanced and summary statistics, such as correlation, fail to capture complexities that lie below the surface. For investors, these complexities matter—accounting for them can make the difference between an effective, or ineffective, hedging strategy. In the case of currencies, investors often determine risk-minimizing hedge ratios based on the portfolio’s betas to those currencies or with mean-variance optimization. In both cases, the optimal solution depends crucially on the correlation between the currencies and assets in the portfolio. But correlation is an unreliable estimate of the diversification investors actually care about: the co-occurrences of the cumulative returns of the portfolio and currencies over the investment horizon. We propose a new currency hedging technique called full-scale hedging, which addresses these challenges by considering the full distribution of co-occurrences between currencies and the portfolio.
By Alexander Cheema-Fox and Robin Greenwood
Using a uniquely deep proprietary dataset, we detail how global investors across regions and asset classes hedge their currency risk, stick to their hedges, and adjust their hedging targets over time.
Currency risk is a key component of global investor returns, but different categories of investor approach these exposures differently. Using State Street’s proprietary custodial data, we have a uniquely precise view into how investors actually choose to hedge and how that varies over time, by asset class, and across different investor domiciles. We introduce a new quantity, the “dynamic hedge ratio,” to capture how investors adjust their hedge ratios and rebalance their currency risk over time. We find that US investors hedge less than others, that equity investors hedge less than fixed-income investors, and that investors tend to stick to target hedge ratios. Moreover, we find that average hedge ratios vary through time with currency, equity, and bond factors, yet exhibit a post GFC shift towards higher hedge ratios that cannot be explained by these factors.
2024 State Street Summer Sessions Webinar Series
This summer we reviewed the fundamentals of finance and investing! Even the most sophisticated investors can benefit from an occasional tune-up. For our fourth annual State Street Summer Sessions, our team of academic and industry experts went back to basics, covering the core principles of modern investing.
Connecting theory to practice, our Global Markets research experts and academic partners covered topics such as inflation, liquidity, private markets, and much more.
CPE credit is offered for those who are CFA charterholders. You can earn 1 hour of credit for attending an hour long Summer Session. To qualify, you must attend the webinar in its entirety, answer the three polling questions throughout the webinar, and submit the Credit Request Form. Please reach out to Insights@StateStreet.com to request your form. We will also send out the form after the webinar. You must specify which session you are requesting credit for.
To view our APAC Region Foundations of Investing Seminar Series, click here.
Thursday June 27, 2024
9am EST
Geopolitics and Markets
Daniel Drezner, Tufts University’s Fletcher School of Law and Diplomacy, State Street Associates Academic Partner | Replay Available
The past few years have highlighted a sea change in how governments approach their own economies and the global economy, adding an additional layer of uncertainty to markets. Geopolitical hotspots have the potential to generate significant economic fallouts. The year of elections is only half over, and the biggest votes are coming soon. Political analyst Daniel Drezner dissects the role that politics will be playing in the months to come.
Tuesday, July 9, 2024
11am EST
Generative AI, Climate Solutions and Investment Implications
George Serafeim, Harvard Business School, State Street Associates Academic Partner| Replay Available
Presenting an application of Generative AI to identify climate technologies and innovation and the implications for growth, risk and valuation across different sectors of the economy.
Thursday, July 11, 2024
9am EST
How Central Banking Relates to Markets and Economies
Robin Greenwood, Harvard Business School, State Street Associates Academic Partner| Replay Available
In an increasingly interconnected world, it is impossible to succeed as an investor without a firm grasp on economic fundamentals and policy levers. In this session, Robin Greenwood ꟷ the George Gund Professor of Finance ꟷ will review the fundamental tenets of central banking with a focus on the main questions global investors should be thinking about in 2024.
Tuesday, July 16, 2024
10am EST
The Limits of Diversification
Will Kinlaw, Senior Managing Director, Head of Global Markets Research, State Street Global Markets | Replay Available
To diversify is one of the fundamental tenets of investing. Yet what seems straightforward in theory is complex in practice. Correlations can be asymmetric and unstable through time. Moreover, correlations measured over shorter intervals do not necessarily extrapolate to longer intervals. This presentation will synthesize more than 10 years of published research into these questions, analyze the challenge from a new perspective, and propose actionable solutions to help investors construct more resilient portfolios.
Thursday, July 18, 2024
10am EST
Inflation Explained: Measurement, Causes, and Latest Trends
Alberto Cavallo, Harvard Business School, State Street Associates Academic Partner| Replay Available
The recent trends indicate that the inflation crisis is ending. However, central banks and investors remain vigilant and cautious about the potential future trajectory, reflecting the persistent uncertainties in the economic landscape. In this session, Alberto Cavallo ꟷ the Thomas S. Murphy Professor of Business Administration at Harvard Business School, co-founder of PriceStats, and member of the Technical Advisory Committee of the U.S. Bureau of Labor Statistics (BLS) ꟷ will discuss the fundamentals of how inflation is measured, what drives it, and how to think about the risk to investors in 2025
Tuesday, July 23, 2024
9am EST
Investing in Private Markets
Josh Lerner, Harvard Business School, State Street Associates Academic Partner| Replay Available
Throughout 2023 and the first half of 2024, private equity faced enormous challenges, navigating lower capital inflow, slower exit activity, decreased valuations and higher capital costs. In this lecture, Harvard Business School professor Josh Lerner will discuss the major factors to consider when investing in today's market conditions.
Professor Lerner will provide insight into the drivers of the historic private equity (PE) boom, current trends that are impacting the direction of the market, and secular shifts that will influence the long-term outlook of PE. The content will draw from a combination of academic research, industry data, and expert insights to provide a 360-degree view of the market landscape. From this lecture, investors will gain a foundation for positioning themselves for success amidst present and future market dynamics.
Thursday July 25, 2024
10am EST
Theory and Practice of Sentiment Analysis Using AI
Gideon Ozik, CFA, PhD, MKT MediaStats, State Street Associates Academic Partner | Replay Available
Analysis of textual information pertaining to stocks, bonds, and currencies can provide investors with valuable insights into market trends and investor behaviors, as well as improve their ability to predict future fluctuations of asset prices.
In this session, we will cover various textual analysis methodologies, review advancements in AI and Large Language Models (LLM), and demonstrate practical applications such as prediction of stock returns using LLMs applied to media coverage, short squeezes using social media, treasury yields using media coverage of monetary policy, and introduce analysis of local media to forecast election outcomes.
Tuesday, July 30, 2024
9am EST
Understanding Chinese Policies and Cross Asset Implications
Ben Luk and Yuting Shao, State Street Global Markets Research | Replay Available
China’s increasing importance not only in emerging markets but also globally means investors are closely following every move out of Beijing. Meanwhile, with China’s post-Covid pent-up demand start to run out of steam, continued weak prices and property sector slump underpin concerns on whether China is able to turn the macro economy around. What’s more, the 3rd Plenum and upcoming US general elections add another layer of policy and geopolitical uncertainty. In this summer session, Ben Luk and Yuting Shao take a deep dive into China’s macro economy and asset classes to try to understand the dynamics of underlying drivers and implications for emerging markets and broader global economy.
Thursday, August 1, 2024
9am EST
Relevance-Based Prediction: A Transparent and Adaptive Alternative to Machine learning
Mark Kritzman, Founding Partner, State Street Associates, State Street Global Markets Founding Partner, CEO, Windham Capital Management, LLC, Chairman, Windham’s Investment Committee | Replay Available
Relevance-based prediction is a model-free approach to prediction that forms predictions as relevance-weighted averages of observed outcomes. The relevance weights are composed of similarity and informativeness, which are both measured as Mahalanobis distances. This prediction method deals with complexities that are beyond the reach of conventional prediction techniques such as linear regression analysis, and it does so in a way that is more transparent, more adaptive, and more theoretically justified than widely used machine learning algorithms.
Tuesday August 6, 2024
9am EST
Quant Strategies and Backtests: Building Blocks and Best Practices
Andrew Li and Alex Cheema-Fox, State Street Associates| Replay Available
We explore the philosophy, mechanisms, and logistics of quantitative strategies and backtesting. This includes how and why to formulate a backtest, modes of testing (e.g. cross-sectional relative value vs market timing), signal construction (simple linear vs machine learning), data wrangling considerations (e.g. ensuring data are point-in-time), and performance evaluation (e.g. risk-adjusted returns, turnover). Illustrative examples from various asset classes are presented.
Thursday, August 8, 2024
9am EST
Addressing Portfolio Risk and Regimes
Meg Czasonis, State Street Associates | Replay Available
Investing always entails risk, and it must be managed. But risk is a multidimensional concept which makes it challenging to measure, and even more challenging to control. In this presentation, Megan Czasonis, head of Portfolio Management Research at State Street Associates, will discuss the benefits and limitations to a range of statistical risk measures—from conventional notions of volatility and value-at-risk to more intricate measurement of losses—as well as conducing regime-specific stress tests and managing portfolio risk.
Tuesday August 13, 2024
10am EST
Machine Learning Interpretation and Model Fingerprint
David Turkington and Huili Song, State Street Associates| Replay Available
Machine learning brings exciting opportunities to investing by utilizing advanced models capable of processing complex nonlinearity and interaction patterns that are powerful for statistical predictions. However, applying machine learning to investing also faces challenges that differ from other disciplines where machine learning has excelled. The primary challenge is the black box problem – the lack of trust and transparency in understanding the models. In this summer session, we will discuss both the opportunities and challenges of applying machine learning to investing, and presenting our solutions that help human users comprehend how a machine learning model arrives at a prediction.
Tuesday, August 20, 2024
11am EST
The Evolution of Crypto Markets
Antoinette Schoar, MIT Sloan School of Management, State Street Associates Academic Partner | Replay Available
Recent developments in the crypto market saw an increasing entry of traditional financial institutions and an expanding role for centralized exchanges. We explore the implications of these trends for systemic risk, data privacy and transparency, as well as consumer financial protection.
Tuesday August 27, 2024
10am EST
Understanding Market Liquidity
Ronnie Sadka, Boston College Carroll School of Management, State Street Associates Academic Partner | Replay Available
Despite having been a key determinant of asset prices for decades, liquidity is still a difficult concept to define and properly understand. In this session, we shall review the theoretical economic underpinnings of market liquidity, and discuss its multi-faceted role in determining market prices and investment strategies. Alternative measures will be introduced as well as practical applications. Further attention will be devoted to the impact of recent market trends, such as retail trading and social media on market liquidity.
Join us for the 2024 State Street Foundations of Investing Seminar Series
This year we reviewed the fundamentals of finance and investing! Even the most sophisticated investors can benefit from an occasional tune-up. For our fourth annual State Street Foundations of Investing Seminar Series, our team of academic and industry experts went back to basics, covering the core principles of modern investing.
Connecting theory to practice, our Global Markets research experts and academic partners covered topics such as inflation, liquidity, private markets, and much more.
To view our NA/EMEA Summer Sessions, click here.
Thursday, July 18, 2024
9am HK
The Limits of Diversification
Will Kinlaw, Senior Managing Director, Head of Global Markets Research, State Street Global Markets | Replay Available
To diversify is one of the fundamental tenets of investing. Yet what seems straightforward in theory is complex in practice. Correlations can be asymmetric and unstable through time. Moreover, correlations measured over shorter intervals do not necessarily extrapolate to longer intervals. This presentation will synthesize more than 10 years of published research into these questions, analyze the challenge from a new perspective, and propose actionable solutions to help investors construct more resilient portfolios.
Thursday, August 1, 2024
9am HK
Understanding Chinese Policies and Cross Asset Implications
Ben Luk and Yuting Shao, State Street Global Markets Research | Replay Available
China’s increasing importance not only in emerging markets but also globally means investors are closely following every move out of Beijing. Meanwhile, with China’s post-Covid pent-up demand start to run out of steam, continued weak prices and property sector slump underpin concerns on whether China is able to turn the macro economy around. What’s more, the 3rd Plenum and upcoming US general elections add another layer of policy and geopolitical uncertainty. In this summer session, Ben Luk and Yuting Shao take a deep dive into China’s macro economy and asset classes to try to understand the dynamics of underlying drivers and implications for emerging markets and broader global economy.
Thursday, August 8, 2024
9am HK
Quant Strategies and Backtests: Building Blocks and Best Practices
Andrew Li and Alex Cheema-Fox, State Street Associates| Replay Available
We explore the philosophy, mechanisms, and logistics of quantitative strategies and backtesting. This includes how and why to formulate a backtest, modes of testing (e.g. cross-sectional relative value vs market timing), signal construction (simple linear vs machine learning), data wrangling considerations (e.g. ensuring data are point-in-time), and performance evaluation (e.g. risk-adjusted returns, turnover). Illustrative examples from various asset classes are presented.
Thursday, August 15, 2024
9am HK
Machine Learning Interpretation and Model Fingerprint
David Turkington and Huili Song, State Street Associates| Replay Available
Machine learning brings exciting opportunities to investing by utilizing advanced models capable of processing complex nonlinearity and interaction patterns that are powerful for statistical predictions. However, applying machine learning to investing also faces challenges that differ from other disciplines where machine learning has excelled. The primary challenge is the black box problem – the lack of trust and transparency in understanding the models. In this summer session, we will discuss both the opportunities and challenges of applying machine learning to investing, and presenting our solutions that help human users comprehend how a machine learning model arrives at a prediction.
Thursday, August 22, 2024
9am HK
Geopolitics and Markets
Daniel Drezner, Tufts University’s Fletcher School of Law and Diplomacy, State Street Associates Academic Partner | Replay Available
The past few years have highlighted a sea change in how governments approach their own economies and the global economy, adding an additional layer of uncertainty to markets. Geopolitical hotspots have the potential to generate significant economic fallouts. The year of elections is only half over, and the biggest votes are coming soon. Political analyst Daniel Drezner dissects the role that politics will be playing in the months to come.
Thursday, August 29, 2024
9am HK
Theory and Practice of Sentiment Analysis Using AI
Gideon Ozik, CFA, PhD, MKT MediaStats, State Street Associates Academic Partner | Replay Available
Analysis of textual information pertaining to stocks, bonds, and currencies can provide investors with valuable insights into market trends and investor behaviors, as well as improve their ability to predict future fluctuations of asset prices.
In this session, we will cover various textual analysis methodologies, review advancements in AI and Large Language Models (LLM), and demonstrate practical applications such as prediction of stock returns using LLMs applied to media coverage, short squeezes using social media, treasury yields using media coverage of monetary policy, and introduce analysis of local media to forecast election outcomes.
State Street Global Markets’ Machine Learning Stock Selection Model uses a data-driven approach and machine learning techniques to predict sector-relative performance of U.S. large cap stocks. Our model:
• Incorporates State Street’s proprietary indicators of media sentiment and behavioral trends
• Uses our Model Fingerprint framework to offer insights into the key drivers of stock returns
• Shows strong efficacy in predicting sector-relative stocks returns in an out-of-sample backtest
The model’s predictions are updated monthly and are available on Insights (click here to learn more). Please note that the Machine Learning Stock Selection Model is currently available for U.S.-based clients.
By Alexander Cheema-Fox, Megan Czasonis, Piyush Kontu, and George Serafeim
We investigate reliance on carbon offsets for decarbonization, associated risks and factors that explain variation in offset prices.
Relying on carbon offsets for decarbonization has become an increasingly contentious approach in the fight against climate change. Many companies view the purchase of credits in carbon reduction or removal projects, such as reforestation or renewable energy initiatives, as an effective way to offset their green house emissions. However, critics argue that offset reliance is a temporary solution that allows for continued emissions rather than addressing the root cause. In a recent paper, we investigate firm reliance on carbon offsets and find that companies tend to use carbon offsets as a complement to their decarbonization activities rather than a substitute. Moreover, we find little evidence that market-based or analyst-derived risk measures reflect the inherent risk associated with offset reliance. Finally, we explore key factors, such as project type and geography, that explain carbon offset quality and prices.
By Megan Czasonis, Mark Kritzman and David Turkington
We show that relevance-based prediction captures complex relationships, like a neural network, but with the added benefit of transparency.
Many prediction tasks in economics and finance lie beyond the reach of linear regression analysis. Researchers, therefore, often turn to machine learning techniques, such as neural networks, to address these complex dynamics. A neural network has the potential to extract nearly all the useful information from a dataset, however it is difficult to implement and notoriously opaque. Alternatively, relevance-based prediction is a model free and theoretically-grounded approach that forms a prediction as a relevance-weighted average of past outcomes. In a sample application to predicting stock market volatility, we show that relevance-based prediction captures complex relationships like a neural network. However, unlike a neural network, it is remarkably transparent, revealing how each observation and variable contributes to a prediction, and disclosing the reliability of a prediction in advance.
By Megan Czasonis, Mark Kritzman and David Turkington
We show how our method of relevance-based prediction implements similar logic to a highly complex machine learning model, but relevance is extremely transparent.
What is the best way to form predictions from a data sample? This is a big question, but at its core lies a fundamental tension between explaining the past and anticipating the future. Predictions can fail by paying too little attention to the past (underfitting) or by paying too much attention (overfitting). High-complexity machine learning models address this problem by recombining past information in thousands (or millions) of exotic ways to map out generalized rules for any situation. An alternative method, called relevance-based prediction, considers each situation one at a time, and extracts the past data that is most useful for that task. We show that there is a deep connection between the two approaches, but only relevance maintains the transparency that makes it easy to explain precisely how each past experience informs a prediction.