Grant: Discovery Launch Supplement
University of Toronto
$12,500.00
Interest Rates and Trading Costs in Equilibrium Models: Stochastic Differential Equations Meet Machine Learning
EF 2024-25
Résumé
An important but only partially answered question in financial economics is why different assets, such as interest rates for safe assets and stock prices for risky assets, earn substantially different returns on average. To address this question, a first-principled path takes the equilibrium modeling approach, which studies the formation of asset pricing models at the fundamental level. With realistic market features, such as fluctuating trading costs and endogenously determined interest rates, equilibrium models are highly nontrivial, since solving the individual portfolio optimization problems for participating investors is already far from myopic, letting alone guarantees of market clearing conditions. The long-term goal of this NSERC Discovery Grant is to landscape equilibrium models with general market features, and incorporate new insights into the design of systematic trading strategies based on crucial market characteristics, in particular for endogenous interest rates and general trading costs. To achieve this long-term goal, my proposed research program will be designed in two short-term research themes: developing analytical machinery to solve equilibrium models with potentially fluctuating trading costs and endogenously determined interest rates, and designing scalable numerical machine learning-based algorithms for equilibrium models with general market dynamics. Through studying several overarching equilibrium models, the analytical Research Theme A will establish mathematical machinery to analyze general equilibrium models. In parallel, the numerical Research Theme B will implement and test several deep reinforcement learning and transfer learning-based numerical algorithms, with insights into the design of a general numerical pipeline that works for equilibrium models with general settings and no longer depends on specific market dynamics.
This proposed research program spans interdisciplinary topics with shared interests in financial engineering, risk management, and machine learning. First, this research will provide mathematical machinery and will obtain quantitative and qualitative behavior of equilibrium systems, serving as cornerstones for future research on financial models with fundamental and realistic setups. Moreover, with calibrations to Nasdaq, NYSE, Euronext and TSE data, it will lead to more accurate modeling and quantitative approximations of crucial market features and better systematic trading strategies. Further, the proposal will provide machine learning-based open-source algorithms, and will significantly increase the computational efficiency and serve as new benchmarks of numerical algorithms, reaching diverse academic and non-academic audiences. With carefully designed HQP training plans and through research, teaching, and leading in my research group, I am fully committed to attracting and educating diverse and competitive researchers and practitioners for Canada's financial and technological sectors.
Programme
Discovery Launch Supplement
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
DGECR-2024-00444
Name nom
Shi, Xiaofei
Department département
Statistical Sciences
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1508
Committee name en
Mathematics and Statistics
Committee name fr
Mathématiques et statistique
Area of application code
1205
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Mathematical sciences
Area of application fr
Sciences mathématiques
Research subject code
3007
Research subject group en
Statistics and probability
Research subject group fr
Statistique et probabilité
Research subject en
Stochastic processes
Research subject fr
Processus stochastiques
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Interest Rates and Trading Costs in Equilibrium Models: Stochastic Differential Equations Meet Machine Learning
Keywords
market microstructure; interest rate models; trading costs; forward-backward stochastic differential equations; equilibrium models; deep reinforcement learning; transfer learning
Application summary
An important but only partially answered question in financial economics is why different assets, such as interest rates for safe assets and stock prices for risky assets, earn substantially different returns on average. To address this question, a first-principled path takes the equilibrium modeling approach, which studies the formation of asset pricing models at the fundamental level. With realistic market features, such as fluctuating trading costs and endogenously determined interest rates, equilibrium models are highly nontrivial, since solving the individual portfolio optimization problems for participating investors is already far from myopic, letting alone guarantees of market clearing conditions. The long-term goal of this NSERC Discovery Grant is to landscape equilibrium models with general market features, and incorporate new insights into the design of systematic trading strategies based on crucial market characteristics, in particular for endogenous interest rates and general trading costs. To achieve this long-term goal, my proposed research program will be designed in two short-term research themes: developing analytical machinery to solve equilibrium models with potentially fluctuating trading costs and endogenously determined interest rates, and designing scalable numerical machine learning-based algorithms for equilibrium models with general market dynamics. Through studying several overarching equilibrium models, the analytical Research Theme A will establish mathematical machinery to analyze general equilibrium models. In parallel, the numerical Research Theme B will implement and test several deep reinforcement learning and transfer learning-based numerical algorithms, with insights into the design of a general numerical pipeline that works for equilibrium models with general settings and no longer depends on specific market dynamics.
This proposed research program spans interdisciplinary topics with shared interests in financial engineering, risk management, and machine learning. First, this research will provide mathematical machinery and will obtain quantitative and qualitative behavior of equilibrium systems, serving as cornerstones for future research on financial models with fundamental and realistic setups. Moreover, with calibrations to Nasdaq, NYSE, Euronext and TSE data, it will lead to more accurate modeling and quantitative approximations of crucial market features and better systematic trading strategies. Further, the proposal will provide machine learning-based open-source algorithms, and will significantly increase the computational efficiency and serve as new benchmarks of numerical algorithms, reaching diverse academic and non-academic audiences. With carefully designed HQP training plans and through research, teaching, and leading in my research group, I am fully committed to attracting and educating diverse and competitive researchers and practitioners for Canada's financial and technological sectors.