Time series forecasting is a key tool in financial markets, helping to predict asset prices and guide investment decisions. In highly volatile markets, such as cryptocurrencies like Bitcoin (BTC) and Ethereum (ETH), forecasting becomes more difficult due to extreme price fluctuations driven by market sentiment, technological changes, and regulatory shifts. Traditionally, forecasting relied on statistical methods, but as markets became more complex, deep learning models like LSTM, Bi-LSTM, and the newer FinBERT-LSTM emerged to capture intricate patterns. Building upon recent advancements and addressing the volatility inherent in cryptocurrency markets, we propose a hybrid model that combines Bidirectional Long Short-Term Memory (Bi-LSTM) networks with FinBERT to enhance forecasting accuracy for these assets. This approach fills a key gap in forecasting volatile financial markets by blending advanced time series models with sentiment analysis, offering valuable insights for investors and analysts navigating unpredictable markets.
Brian Z. Zhu, Dingyue Liu, Xin Wan, Gordon Liao · 6 authors
We study liquidity on decentralized exchanges (DEXs), identifying factors at the platform, blockchain, token pair, and liquidity pool levels with predictive power for market depth metrics. We introduce the v2 counterfactual spread metric, a novel criterion which assesses the degree of liquidity concentration in pools using the ``concentrated liquidity'' mechanism, allowing us to decompose the effect of a factor on market depth into two channels: total value locked (TVL) and concentration. We further explore how external liquidity from competing DEXs and private inventory on DEX aggregators influence market depth. We find that (i) gas prices, returns, and a DEX's share of trading volume affect liquidity through concentration, (ii) internalization of order flow by private market makers affects TVL but not the overall market depth, and (iii) volatility, fee revenue, and markout affect liquidity through both channels.
This study utilizes machine learning algorithms to analyze and organize knowledge in the field of algorithmic trading. By filtering a dataset of 136 million research papers, we identified 14,342 relevant articles published between 1956 and Q1 2020. We compare traditional practices-such as keyword-based algorithms and embedding techniques-with state-of-the-art topic modeling methods that employ dimensionality reduction and clustering. This comparison allows us to assess the popularity and evolution of different approaches and themes within algorithmic trading. We demonstrate the usefulness of Natural Language Processing (NLP) in the automatic extraction of knowledge, highlighting the new possibilities created by the latest iterations of Large Language Models (LLMs) like ChatGPT. The rationale for focusing on this topic stems from our analysis, which reveals that research articles on algorithmic trading are increasing at a faster rate than the overall number of publications. While stocks and main indices comprise more than half of all assets considered, certain asset classes, such as cryptocurrencies, exhibit a much stronger growth trend. Machine learning models have become the most popular methods in recent years. The study demonstrates the efficacy of LLMs in refining datasets and addressing intricate questions about the analyzed articles, such as comparing the efficiency of different models. Our research shows that by decomposing tasks into smaller components and incorporating reasoning steps, we can effectively tackle complex questions supported by case analyses. This approach contributes to a deeper understanding of algorithmic trading methodologies and underscores the potential of advanced NLP techniques in literature reviews.
We provide an economic model of Execution Tickets and use it to study the ability of the Ethereum protocol to capture MEV from block construction. We demonstrate that Execution Tickets extract all MEV when all buyers are homogeneous, risk neutral and face no capital costs. We also show that MEV capture decreases with risk aversion and capital costs. Moreover, when buyers are heterogeneous, MEV capture can be especially low and a single dominant buyer can extract much of the MEV. This adverse effect can be partially mitigated by the presence of a Proposer Builder Separation (PBS) mechanism, which gives ET buyers access to a market of specialized builders, but in practice centralization vectors still persist. With PBS, ETs are concentrated among those with the highest ex-ante MEV extraction ability and lowest cost of capital. We show how it is possible that large investors that are not builders but have substantial advantage in capital cost can come to dominate the ET market.
Rob McLaughlin, Nir Chemaya, Dingyue Liu, Dahlia Malkhi
This paper introduces a trade ordering rule that aims to reduce intra-block price volatility in Automated Market Maker (AMM) powered decentralized exchanges. The ordering rule introduced here, Clever Look-ahead Volatility Reduction (CLVR), operates under the (common) framework in decentralized finance that allows some entities to observe trade requests before they are settled, assemble them into "blocks", and order them as they like. On AMM exchanges, asset prices are continuously and transparently updated as a result of each trade and therefore, transaction order has high financial value. CLVR aims to order transactions for traders' benefit. Our primary focus is intra-block price stability (minimizing volatility), which has two main benefits for traders: it reduces transaction failure rate and allows traders to receive closer prices to the reference price at which they submit their transactions accordingly. We show that CLVR constructs an ordering which approximately minimizes price volatility with a small computation cost and can be trivially verified externally.
Automated market makers (AMMs) are a new type of trading venues which are revolutionising the way market participants interact. At present, the majority of AMMs are constant function market makers (CFMMs) where a deterministic trading function determines how markets are cleared. Within CFMMs, we focus on constant product market makers (CPMMs) which implements the concentrated liquidity (CL) feature. In this thesis we formalise and study the trading mechanism of CPMMs with CL, and we develop liquidity provision and liquidity taking strategies. Our models are motivated and tested with market data. We derive optimal strategies for liquidity takers (LTs) who trade orders of large size and execute statistical arbitrages. First, we consider an LT who trades in a CPMM with CL and uses the dynamics of prices in competing venues as market signals. We use Uniswap v3 data to study price, liquidity, and trading cost dynamics, and to motivate the model. Next, we consider an LT who trades a basket of crypto-currencies whose constituents co-move. We use market data to study lead-lag effects, spillover effects, and causality between trading venues. We derive optimal strategies for strategic liquidity providers (LPs) who provide liquidity in CPMM with CL. First, we use stochastic control tools to derive a self-financing and closed-form optimal liquidity provision strategy where the width of the LP's liquidity range is determined by the profitability of the pool, the dynamics of the LP's position, and concentration risk. Next, we use a model-free approach to solve the problem of an LP who provides liquidity in multiple CPMMs with CL. We do not specify a model for the stochastic processes observed by LPs, and use a long short-term memory (LSTM) neural network to approximate the optimal liquidity provision strategy.
Cryptocurrency is a cryptography-based digital asset with extremely volatile prices. Around USD 70 billion worth of cryptocurrency is traded daily on exchanges. Trading cryptocurrency is difficult due to the inherent volatility of the crypto market. This study investigates whether Reinforcement Learning (RL) can enhance decision-making in cryptocurrency algorithmic trading compared to traditional methods. In order to address this question, we combined reinforcement learning with a statistical arbitrage trading technique, pair trading, which exploits the price difference between statistically correlated assets. We constructed RL environments and trained RL agents to determine when and how to trade pairs of cryptocurrencies. We developed new reward shaping and observation/action spaces for reinforcement learning. We performed experiments with the developed reinforcement learner on pairs of BTC-GBP and BTC-EUR data separated by 1 min intervals (n=263,520). The traditional non-RL pair trading technique achieved an annualized profit of 8.33%, while the proposed RL-based pair trading technique achieved annualized profits from 9.94% to 31.53%, depending upon the RL learner. Our results show that RL can significantly outperform manual and traditional pair trading techniques when applied to volatile markets such as~cryptocurrencies.
The year 2024 witnessed a major development in the cryptocurrency industry with the long-awaited approval of spot Bitcoin exchange-traded funds (ETFs). This innovation provides investors with a new, regulated path to gain exposure to Bitcoin through a familiar investment vehicle (Kumar et al., 2024). However, unlike traditional ETFs that directly hold underlying assets, Bitcoin ETFs rely on a creation and redemption process managed by authorized participants (APs). This unique structure introduces distinct characteristics in terms of premium/discount behavior compared to traditional ETFs. This paper investigates the premium and discount patterns observed in Bitcoin ETFs during first four-month period (January 11th, 2024, to May 17th, 2024). Our analysis reveals that these patterns differ significantly from those observed in traditional index ETFs, potentially exposing investors to additional risk factors. By identifying and analyzing these risk factors associated with Bitcoin ETF premiums/discounts, this paper aims to achieve two key objectives: · Enhance market understanding: Equip and market and investors with a deeper comprehension of the unique liquidity risks inherent in Bitcoin ETFs. · Provide a clearer risk management frameworks: Offer a clearer perspective on the risk-return profile of digital asset ETFs, specifically focusing on Bitcoin ETFs. Through a thorough analysis of premium/discount behavior and the underlying factors contributing to it, this paper strives to contribute valuable insights for investors navigating the evolving landscape of digital asset investments
Hongzhou Chen, Xiaolin Duan, Abdulmotaleb El Saddik, Wei Cai
Harnessing the transparent blockchain user behavior data, we construct the Political Betting Leaning Score (PBLS) to measure political leanings based on betting within Web3 prediction markets. Focusing on Polymarket and starting from the 2024 U.S. Presidential Election, we synthesize behaviors over 15,000 addresses across 4,500 events and 8,500 markets, capturing the intensity and direction of their political leanings by the PBLS. We validate the PBLS through internal consistency checks and external comparisons. We uncover relationships between our PBLS and betting behaviors through over 800 features capturing various behavioral aspects. A case study of the 2022 U.S. Senate election further demonstrates the ability of our measurement while decoding the dynamic interaction between political and profitable motives. Our findings contribute to understanding decision-making in decentralized markets, enhancing the analysis of behaviors within Web3 prediction environments. The insights of this study reveal the potential of blockchain in enabling innovative, multidisciplinary studies and could inform the development of more effective online prediction markets, improve the accuracy of forecast, and help the design and optimization of platform mechanisms. The data and code for the paper are accessible at the following link: https://github.com/anonymous.
Riccardo De Blasis, Luca Galati, Rosanna Grassi, Giorgio Rizzini
This paper investigates the cryptocurrency network of the FTX exchange during the collapse of its native token, FTT, to understand how network structures adapt to significant financial disruptions, by exploiting vertex centrality measures. Using proprietary data on the transactional relationships between various cryptocurrencies, we construct the filtered correlation matrix to identify the most significant relations in the FTX and Binance markets. By using suitable centrality measures - closeness and information centrality - we assess network stability during FTX's bankruptcy. The findings document the appropriateness of such vertex centralities in understanding the resilience and vulnerabilities of financial networks. By tracking the changes in centrality values before and during the FTX crisis, this study provides useful insights into the structural dynamics of the cryptocurrency market. Results reveal how different cryptocurrencies experienced shifts in their network roles due to the crisis. Moreover, our findings highlight the interconnectedness of cryptocurrency markets and how the failure of a single entity can lead to widespread repercussions that destabilize other nodes of the network.
This study evaluates the performance of 41 machine learning models, including 21 classifiers and 20 regressors, in predicting Bitcoin prices for algorithmic trading. By examining these models under various market conditions, we highlight their accuracy, robustness, and adaptability to the volatile cryptocurrency market. Our comprehensive analysis reveals the strengths and limitations of each model, providing critical insights for developing effective trading strategies. We employ both machine learning metrics (e.g., Mean Absolute Error, Root Mean Squared Error) and trading metrics (e.g., Profit and Loss percentage, Sharpe Ratio) to assess model performance. Our evaluation includes backtesting on historical data, forward testing on recent unseen data, and real-world trading scenarios, ensuring the robustness and practical applicability of our models. Key findings demonstrate that certain models, such as Random Forest and Stochastic Gradient Descent, outperform others in terms of profit and risk management. These insights offer valuable guidance for traders and researchers aiming to leverage machine learning for cryptocurrency trading.
Current digital payment solutions are fragile and offer less privacy than traditional cash. Their critical dependency on an online service used to perform and validate transactions makes them void if this service is unreachable. Moreover, no transaction can be executed during server malfunctions or power outages. Due to climate change, the likelihood of extreme weather increases. As extreme weather is a major cause of power outages, the frequency of power outages is expected to increase. The lack of privacy is an inherent result of their account-based design or the use of a public ledger. The critical dependency and lack of privacy can be resolved with a Central Bank Digital Currency that can be used offline. This thesis proposes a design and a first implementation for an offline-first digital euro. The protocol offers complete privacy during transactions using zero-knowledge proofs. Furthermore, transactions can be executed offline without third parties and retroactive double-spending detection is facilitated. To protect the users' privacy, but also guard against money laundering, we have added the following privacy-guarding mechanism. The bank and trusted third parties for law enforcement must collaborate to decrypt transactions, revealing the digital pseudonym used in the transaction. Importantly, the transaction can be decrypted without decrypting prior transactions attached to the digital euro. The protocol has a working initial implementation showcasing its usability and demonstrating functionality.
The utilization of Large Language Models (LLMs) in financial trading has primarily been concentrated within the stock market, aiding in economic and financial decisions. Yet, the unique opportunities presented by the cryptocurrency market, noted for its on-chain data's transparency and the critical influence of off-chain signals like news, remain largely untapped by LLMs. This work aims to bridge the gap by developing an LLM-based trading agent, CryptoTrade, which uniquely combines the analysis of on-chain and off-chain data. This approach leverages the transparency and immutability of on-chain data, as well as the timeliness and influence of off-chain signals, providing a comprehensive overview of the cryptocurrency market. CryptoTrade incorporates a reflective mechanism specifically engineered to refine its daily trading decisions by analyzing the outcomes of prior trading decisions. This research makes two significant contributions. Firstly, it broadens the applicability of LLMs to the domain of cryptocurrency trading. Secondly, it establishes a benchmark for cryptocurrency trading strategies. Through extensive experiments, CryptoTrade has demonstrated superior performance in maximizing returns compared to traditional trading strategies and time-series baselines across various cryptocurrencies and market conditions. Our code and data are available at \url{https://anonymous.4open.science/r/CryptoTrade-Public-92FC/}.
High-frequency trading (HFT) that executes algorithmic trading in short time scales, has recently occupied the majority of cryptocurrency market. Besides traditional quantitative trading methods, reinforcement learning (RL) has become another appealing approach for HFT due to its terrific ability of handling high-dimensional financial data and solving sophisticated sequential decision-making problems, \emph{e.g.,} hierarchical reinforcement learning (HRL) has shown its promising performance on second-level HFT by training a router to select only one sub-agent from the agent pool to execute the current transaction. However, existing RL methods for HFT still have some defects: 1) standard RL-based trading agents suffer from the overfitting issue, preventing them from making effective policy adjustments based on financial context; 2) due to the rapid changes in market conditions, investment decisions made by an individual agent are usually one-sided and highly biased, which might lead to significant loss in extreme markets. To tackle these problems, we propose a novel Memory Augmented Context-aware Reinforcement learning method On HFT, \emph{a.k.a.} MacroHFT, which consists of two training phases: 1) we first train multiple types of sub-agents with the market data decomposed according to various financial indicators, specifically market trend and volatility, where each agent owns a conditional adapter to adjust its trading policy according to market conditions; 2) then we train a hyper-agent to mix the decisions from these sub-agents and output a consistently profitable meta-policy to handle rapid market fluctuations, equipped with a memory mechanism to enhance the capability of decision-making. Extensive experiments on various cryptocurrency markets demonstrate that MacroHFT can achieve state-of-the-art performance on minute-level trading tasks.
Milionis et al.(2023) studied the rate at which automated market makers leak value to arbitrageurs when block times are discrete and follow a Poisson process, and where the risky asset price follows a geometric Brownian motion. We extend their model to analyze another popular mechanism in decentralized finance for onchain trading: Dutch auctions. We compute the expected losses that a seller incurs to arbitrageurs and expected time-to-fill for Dutch auctions as a function of starting price, volatility, decay rate, and average interblock time. We also extend the analysis to gradual Dutch auctions, a variation on Dutch auctions for selling tokens over time at a continuous rate. We use these models to explore the tradeoff between speed of execution and quality of execution, which could help inform practitioners in setting parameters for starting price and decay rate on Dutch auctions, or help platform designers determine performance parameters like block times.
We have designed an innovative portfolio rebalancing mechanism termed the Cascading Waterfall Round Robin Mechanism. This algorithmic approach recommends an ideal size and number of trades for each asset during the periodic rebalancing process, factoring in the gas fee and slippage. The essence of the model we have created gives indications regarding whether trades should be made on individual assets depending on the uncertainty in the micro - asset level characteristics - and macro - aggregate market factors - environments. In the hyper-volatile crypto market, our approach to daily rebalancing will benefit from volatility. Price movements will cause our algorithm to buy assets that drop in prices and sell as they soar. In fact, the buying and selling happen only when certain boundaries are crossed in order to weed out any market noise and ensure sound trade execution. We have provided several numerical examples to illustrate the steps - including the calculation of several intermediate variables - of our rebalancing mechanism. The Algorithm we have developed can be easily applied outside blockchain to investment funds across all asset classes at any trading frequency and rebalancing duration. Shakespeare As A Crypto Trader: To Trade Or Not To Trade, that is the Question, Whether an Optimizer can Yield the Answer, Against the Spikes and Crashes of Markets Gone Wild, To Quench One's Thirst before Liquidity Runs Dry, Or Wait till the Tide of Momentum turns Mild.
This work introduces a framework for evaluating onchain order flow auctions (OFAs), emphasizing the metric of price improvement. Utilizing a set of open-source tools, our methodology systematically attributes price improvements to specific modifiable inputs of the system such as routing efficiency, gas optimization, and priority fee settings. When applied to leading Ethereum-based trading interfaces such as 1Inch and Uniswap, the results reveal that auction-enhanced interfaces can provide statistically significant improvements in trading outcomes, averaging 4-5 basis points in our sample. We further identify the sources of such price improvements to be added liquidity for large swaps. This research lays a foundation for future innovations in blockchain based trading platforms.
We investigate the behavior of liquidity providers (LPs) by modeling a decentralized cryptocurrency exchange (DEX) based on Uniswap v3. LPs with heterogeneous characteristics choose optimal liquidity positions subject to uncertainty regarding the size of exogenous incoming transactions and the prices of assets in the wider market. They engage in a game among themselves, and the resulting liquidity distribution determines the exchange rate dynamics and potential arbitrage opportunities of the pool. We calibrate the distribution of LP characteristics based on Uniswap data and the equilibrium strategy resulting from this mean-field game produces pool exchange rate dynamics and liquidity evolution consistent with observed pool behavior. We subsequently introduce Maximal Extractable Value (MEV) bots who perform Just-In-Time (JIT) liquidity attacks, and develop a Stackelberg game between LPs and bots. This addition results in more accurate simulated pool exchange rate dynamics and stronger predictive power regarding the evolution of the pool liquidity distribution.
This paper presents the results of a comprehensive empirical study of losses to arbitrageurs (following the formalization of loss-versus-rebalancing by [Milionis et al., 2022]) incurred by liquidity providers on automated market makers (AMMs). We show that those losses exceed the fees earned by liquidity providers across many of the largest AMM liquidity pools (on Uniswap). Remarkably, we also find that the Uniswap v2 pools are more profitable for passive LPs than their Uniswap v3 counterparts. We also investigate how arbitrage losses change with block times. As expected, arbitrage losses decrease when block production is faster. However, the rate of the decline varies significantly across different trading pairs. For instance, when comparing 100ms block times to Ethereum's current 12-second block times, the decrease in losses to arbitrageurs ranges between 20% to 70%, depending on the specific trading pair.
We study how trading fees and continuous-time arbitrage affect the profitability of liquidity providers (LPs) in Geometric Mean Market Makers (G3Ms). We use stochastic reflected diffusion processes to analyze the dynamics of a G3M model under the arbitrage-driven market [Milionis et al. 2022a. “Automated Market Making and Loss-Versus-Rebalancing.” arXiv e-prints]. Our research focuses on calculating LP wealth and extends the findings of Tassy and White [Tassy and White. 2020. “Growth Rate of a Liquidity Provider's Wealth in xy = c Automated Market Makers.”] for the constant product market maker (Uniswap v2) to a broader range of G3Ms, including Balancer. This allows us to calculate the long-term expected logarithmic growth of LP wealth, offering new insights into the complex dynamics of AMMs and their implications for LPs in decentralized finance.
National football teams increasingly issue tradeable blockchain-based fan tokens to strategically enhance fan engagement. This study investigates the impact of 2022 World Cup matches on the dynamic performance of each team's fan token. The event study uncovers fan token returns surged six months before the World Cup, driven by positive anticipation effects. However, intraday analysis reveals a reversal of fan token returns consistently declining and trading volumes rising as matches unfold. To explain findings, we uncover asymmetries whereby defeats in high-stake matches caused a plunge in fan token returns, compared to low-stake matches, intensifying in magnitude for knockout matches. Contrarily, victories enhance trading volumes, reflecting increased market activity without a corresponding positive effect on returns. We align findings with the classic market adage "buy the rumor, sell the news," unveiling cognitive biases and nuances in investor sentiment, cautioning the dichotomy of pre-event optimism and subsequent performance declines.
This paper studies the market structure impact of cheaper and faster chains on the Uniswap v3 Protocol. The Uniswap Protocol is the largest decentralized application on Ethereum by both gas and blockspace used, and user behaviors of the protocol are very sensitive to fluctuations in gas prices and market structure due to the economic factors of the Protocol. We focus on the chains where Uniswap v3 has the most activity, giving us the best comparison to Ethereum mainnet. Because of cheaper gas and lower block times, we find evidence that the majority of swaps get better gas-adjusted execution on these chains, liquidity providers are more capital efficient, and liquidity providers have increased fee returns from more arbitrage. We also present evidence that two second block times may be too long for optimal liquidity provider returns, compared to first come, first served. We argue that many of the current drawbacks with AMMs may be due to chain dynamics and are vastly improved with cheaper and faster transactions
We develop several innovations to bring the best practices of traditional investment funds to the blockchain landscape. Specifically, we illustrate how: 1) fund prices can be updated regularly like mutual funds; 2) performance fees can be charged like hedge funds; 3) mutually hedged blockchain investment funds can operate with investor protection schemes, such as high water marks; and 4) measures to offset trading related slippage costs when redemptions happen. Using our concepts - and blockchain technology - traditional funds can calculate performance fees in a simplified manner and alleviate several operational issues. Blockchain can solve many problems for traditional finance, while tried and tested wealth management techniques can benefit decentralization, speeding its adoption. We provide detailed steps - including mathematical formulations and instructive pointers - to implement these ideas and discuss how our designs overcome several blockchain bottlenecks, making smart contracts smarter. We provide numerical illustrations of several scenarios related to our mechanisms.
Austin Adams, Ciamac C. Moallemi, Sara Reynolds, Dan Robinson
Automated market makers (AMMs) have emerged as the dominant market mechanism for trading on decentralized exchanges implemented on blockchains. This paper presents a single mechanism that targets two important unsolved problems for AMMs: reducing losses to informed orderflow, and maximizing revenue from uninformed orderflow. The ``auction-managed AMM'' works by running a censorship-resistant onchain auction for the right to temporarily act as ``pool manager'' for a constant-product AMM. The pool manager sets the swap fee rate on the pool, and also receives the accrued fees from swaps. The pool manager can exclusively capture some arbitrage by trading against the pool in response to small price movements, and also can set swap fees incorporating price sensitivity of retail orderflow and adapting to changing market conditions, with the benefits from both ultimately accruing to liquidity providers. Liquidity providers can enter and exit the pool freely in response to changing rent, though they must pay a small fee on withdrawal. We prove that under certain assumptions, this AMM should have higher liquidity in equilibrium than any standard, fixed-fee AMM.