Grant ID: 861mttufj
Summary:
Motivation
Yield farming is the practice of seeking rewards in the form of transaction fees by depositing tokens into a liquidity pool on a decentralized application (dApp). In contrast to traditional finance, yield farming is different from traditional liquidity provision because it occurs in the absence of a central market maker (intermediary). If a trader (the liquidity taker) enters the liquidity pool, he can buy and sell tokens by swapping them in the pool according to a known pricing rule governed by the Automated Market Maker (AMM). In exchange for providing liquidity, the yield farmer (liquidity provider) earns a share of the trading fees from the trades that occur in the liquidity pool. Liquidity is of critical importance to the stability and efficiency of financial markets. In yield farming, tokens are lent (to a protocol), and, most likely, the protocol will not return the same quantities of individual tokens to the lender, due to the so-called impermanent loss.
Impermanent loss defines the opportunity-cost dynamic between offering assets for exchange and holding the underlying assets to potentially profit from the price movement. This occurs when the price of the deposited assets changes compared to when one deposited them. The bigger this change is, the more one is exposed to impermanent loss. It is important to mention that an impermanent loss does not necessarily mean that the LP experiences a negative return on the investment: It simply means that the gains from a buy-and-hold strategy outperformed the returns obtained for the liquidity provision.
Contribution
The academic literature on impermanent loss as the major risk driver in DeFi is not developed yet and especially little is known about how to hedge or minimize the impermanent loss for investors.
The main contributions we plan to achieve in this project are as follows:
i) To shed light on the concrete mechanism of impermanent loss, we propose a continuous-time stochastic model which we calibrate to the data and then use to characterize the impermanent loss dynamics in the liquidity pool. The construction of our model starts with the mathematical formulation of the AMM governing prices in the liquidity pool. We then fully derive and characterize the impermanent loss as the major source of risk for the agent who has invested in the pool. For a single pool with a single pair of tokens, the first calculations show that the impermanent loss is a function of the respective token return volatilities and the correlation between them.
ii) We then study the replication (hedging) of the impermanent loss: Therefore, we solve a portfolio problem using stochastic control theory with the goal to minimize the (one-step ahead) impermanent loss occurring to the yield farmer. The optimal portfolio strategy will then command the optimal holdings in either the underlying token pair (outside the pool) or derivatives (call or put options) written on the respective token individually.
I have read and agree to the privacy policy:
true
Describe what your project does or is aiming to do in 50 characters or less:
Academic Project, The Impermanent Loss (IL) in Yield Farming; Continuous-time Stochastic Model, Hedging of the IL via tokens and derivatives
How far along are you?:
We started with explanatory work and a detailed literature review. Currently, we are seeking grants to finance this project.
Why did you pick this idea to work on? Do you have domain expertise in this area? How do you know people need what you’re making?:
Yes, we have expertise in this research area: Our current paper “Maneuvering and Investing in Yield Farms” (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4422213) got presented already at various conferences.
The Team:
Prof. Dr. Lorenzo Schoenleber is an Assistant Professor in Finance at the Collegio Carlo Alberto and the University of Turin. He obtained his PhD at the Frankfurt School of Finance. He is also associated with the Fintech & Digital Finance Chair at Paris Dauphine University. His area of specialization is empirical asset pricing (option-implied information) and DeFi (Yield Farming).
Prof. Dr. Andrew Papanicolaou is an associate professor in the Department of Mathematics at North Carolina State University (NCSU). His PhD is in applied mathematics from Brown University. His research interests are computational finance and stochastic systems for control and optimization. The applications of this work include financial data analysis and the challenges associated with these highly complex data sets. My background is in probability theory and nonlinear filtering.
Dr. Siddharth Naik is a quantitative portfolio manager and trader with over 7 years of experience at institutional hedge funds (systematic macro and low latency, respectively) with extensive experience in trading fx, futures, and crypto instruments. He has previously published multiple papers on non-convex optimization mechanism design and trading crypto since 2017. Previously he was a CEO at RadioLytics a startup, which aimed at making spectrum tradable.
What’s new about what you’re making? What substitutes do people resort to because it doesn’t exist yet (or they don’t know about it)?:
The academic literature on impermanent loss as the major risk driver in DeFi is not developed yet and especially little is known about how to hedge or minimize the impermanent loss for investors.
What’s your vision for how your project will impact The Graph and/or web3?:
The stability and efficiency of financial markets heavily rely on the crucial significance of liquidity. You would sponsor a high-level academic research paper in DeFi and hence gain visibility among the academic community. Our current paper “Maneuvering and Investing in Yield Farms” (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4422213) got presented already at various conferences. Besides the visibility, the results which we provide in our paper can be useful for the enhancement and development of the current architecture of your dApp. For example, via a simulation study, one could provide corridors for the expected impermanent loss. Besides, the paper would guide and educate the investor on how to avoid or hedge impermanent loss.
In addition, we just held a one-day conference by the Collegio Carlo Alberto and the University of Turin, called ToDeFi: Torino Decentralized Finance Conference 2023 (ToDeFi: Torino Decentralized Finance Conference 2023 | Collegio Carlo Alberto), which promoted the highest level and up-to-date research in DeFi within the academic community. Speakers and discussants were selected among the most innovative junior researchers and established scholars. We are planning to make the conference a yearly event. Beyond the research support, we are potentially open to discuss about a deeper cooperation. In particular, the Collegio Carlo Alberto hosts a postgraduate Master’s in finance, insurance, and risk management (Mafirm), which is a small high-quality program that exists for more than 20 years. The program appears in several international rankings (e.g., 1st in Italy, 6th in Europe, and 25th worldwide in the quantitative finance masters’ ranking by Risk.net). The students attending our master lectures are excellent but more importantly, really motivated.
How will you measure success?:
Measuring success for a grant related to an academic paper can be assessed using various criteria.
We aim for the successful completion of the academic paper and its subsequent publication in a reputable journal. The publication demonstrates that the research was conducted effectively and contributes to the academic community. In addition, the impact of the paper can be measured by the number of citations which suggests that the research has been influential and has generated interest within the academic community. Success can also be evaluated based on the opportunities for collaborations and networking that arise from the research (such as our yearly conference ToDeFi). If the grant helps establish connections with other researchers, institutions, or organizations, it can enhance the visibility and impact of the work. In addition, we are aim for research which is practically applicable having a real-world impact. If the findings of the paper are implemented in practice or contribute to solving real-world problems, it demonstrates the relevance and value of the research.
What milestones will you achieve in the coming weeks/months, with or without the grant?:
06-08/2023 – Accessing, preparing, and processing spot and option data via the Tardis API
08-09/2023 – Accessing liquidity pool data using The Graph
09-12/2023 – Developing the mathematical optimization framework
01-02/2024 – Programming and numerically solving the optimization framework
02-04/2024 – Calibrating the model output to the data (tokens and options)
04-06/2024 – First paper draft
07-08/2024 – Submission to a Journal
Please select who you think will benefit from your project?:
Other
How much funding are you looking for? (USD):
3000
Provide a breakdown of your requested funding:
In order to implement our optimization framework, we have to model the underlying relevant stochastic processes: While current spot prices are publicly available, dedicated firms such as https://tardis.dev offer access to various cryptocurrency derivative trading platforms, which we need to acquire for this project. Pool data can be sourced for free via The Graph as done in our current project. On top of the data acquisition, we need to hire a Research Assistant, who helps us to access and analyze the relevant data. The optimization framework requires high-level hardware and computing capabilities which we need to acquire (or rent) for this project. The price for the Tardis API for 1 year is about 6.000€. Hiring a Research Assistant for about 1.5 years accumulates to approximately 8000€. Hardware can be acquired for less than 2000€.
We want to emphasize that any amount of sponsorship or donation would be beneficial for us since we are trying to request funding from different funding sources to realize this project.
Have you applied for (or already received) funding from somewhere else?:
Yes
If yes, please describe where you’ve received funding from and how much you have, or will, receive:
We haven’t received funding yet but we are trying to request funding from different funding sources to realize this project.
Can you share any information about current/past web3 experience?:
Prof. Dr. Lorenzo Schoenleber is an Assistant Professor in Finance at the Collegio Carlo Alberto and the University of Turin. He obtained his PhD at the Frankfurt School of Finance. He is also associated with the Fintech & Digital Finance Chair at Paris Dauphine University. His area of specialization is empirical asset pricing (option-implied information) and DeFi (Yield Farming).
Prof. Dr. Andrew Papanicolaou is an associate professor in the Department of Mathematics at North Carolina State University (NCSU). His PhD is in applied mathematics from Brown University. His research interests are computational finance and stochastic systems for control and optimization. The applications of this work include financial data analysis and the challenges associated with these highly complex data sets. My background is in probability theory and nonlinear filtering.
Dr. Siddharth Naik is a quantitative portfolio manager and trader with over 7 years of experience at institutional hedge funds (systematic macro and low latency, respectively) with extensive experience in trading fx, futures, and crypto instruments. He has previously published multiple papers on non-convex optimization mechanism design and trading crypto since 2017. Previously he was a CEO at RadioLytics a startup, which aimed at making spectrum tradable.
Why are you interested in contributing to The Graph and/or web3?:
We want to contribute to DeFi and Web3 because the academic literature on impermanent loss as the major risk driver in DeFi is not developed yet and especially little is known about how to hedge or minimize the impermanent loss for investors.
Are you applying as a team or individual?:
Team
What type of grant are you pursuing?:
Project (one-off)
Please select the category your project best fits into:
Other
Primary Community Poll
- Yes
- No
0 voters
Grant Committee Vote
- Yes
- No
0 voters