This study revisits calendar anomalies in cryptocurrency markets using hourly data for four actively traded cryptocurrencies (Bitcoin, Ethereum, Tether USDt, and BNB) and eight less active cryptocurrencies. While prior studies based on daily returns provide mixed evidence on day-of-the-week (DoW) effects, we show that these patterns are not persistent daily phenomena. Instead, they are driven by a limited number of intraday intervals and do not reflect broad daily behavior.We further document that these effects are localized, asset-specific, and more pronounced among actively traded cryptocurrencies, while largely absent among less active ones.Overall, the findings indicate that cryptocurrency markets exhibit limited and short-lived inefficiencies rather than persistent anomalies. This highlights the importance of employing high-frequency data in studies of continuously traded markets and suggests that digital asset investors may benefit from the presence of abnormal returns only in a limited number of hours on specific days of the week.
This paper documents a robust size-dependent pattern in cryptocurrency return predictability. Small coins exhibit strong short-term reversal, whereas large coins exhibit momentum, and the relation varies monotonically across the size distribution. We further show that these two sides of the pattern reflect different return dynamics: small-coin reversal is driven mainly by rebounds among recent losers, while large-coin momentum reflects their continued underperformance. Liquidity frictions and idiosyncratic volatility explain part of this pattern, but not all of it. These findings point to a "size-momentum puzzle" in cryptocurrencies.
This study develops and tests a theoretical framework linking liquidity, market integration, and return predictability in cryptocurrency markets. Analyzing high-frequency daily data for five major cryptocurrencies (Bitcoin, Ethereum, Binance Coin, Solana, and Ripple), we employ Random Forest models augmented with rigorous time-series diagnostics and economic significance tests. Our findings establish a fundamental dichotomy: Bitcoin exhibits predictability driven by macroeconomic fundamentals (global risk-free rates, economic policy uncertainty), consistent with its emergence as a macro-asset; altcoins, in contrast, are dominated by internal microstructure (realized volatility, illiquidity) and speculative sentiment. Formal hypothesis tests confirm that (i) lower liquidity predicts higher future returns, (ii) machine learning models systematically underpredict during positively skewed regimes, and (iii) macro integration attenuates microstructure-driven predictability. Out-of-sample R 2 values reach 6.8% for Bitcoin and 9.6% for Ethereum, with Diebold-Mariano statistics rejecting equal predictive accuracy against a random walk at the 1% level. A mean-variance investor would earn a certainty equivalent gain of 2.4% annually by exploiting these forecasts. The results challenge the efficient market hypothesis for 1 digital assets, establish a new taxonomy of cryptocurrency predictability, and provide critical implications for asset pricing, portfolio allocation, and risk management in decentralized finance.
Predicting short-term mid-price movements from limit order book (LOB) data is a fundamental problem in quantitative finance and market microstructure research, with direct applicability to both traditional exchanges and cryptocurrency marketsâincluding centralized exchanges (CEXs) and emerging on-chain LOB protocols in decentralized finance (DeFi). We present three contributions to this domain. First, we propose DA-BiGRU-CNN, a domain-aware dual-branch architecture that decomposes LOB features into price and volume information channels, processes them through dedicated bidirectional GRU encoders with shared microstructure features, and fuses temporal representations via a multi-scale convolutional bottleneck (Conv1d with kernels k = 3,5,7). Second, we provide empirical evidence for a "feature sufficiency" hypothesis: a unidirectional GRU trained on 53 basic features achieves performance statistically equivalent to one trained on 219 extensively engineered featuresâincluding rolling statistics, exponential moving averages, and lag/difference featuresâsuggesting that recurrent hidden states implicitly learn these temporal patterns. Third, we document a "negative ensemble effect" where combining sequential (GRU) and tabular (gradient boosting) models consistently degrades prediction quality, contradicting the widely-held assumption that model diversity improves ensemble performance. On a large-scale dataset of 12,165 LOB sequences (12.1M timesteps), our GRU baseline achieves a weighted Pearson correlation of 0.266, outperforming LightGBM by 58%, while our domain-aware architecture offers an architecturally principled alternative that naturally separates price dynamics from liquidity dynamics.
We document significant reversal in cryptocurrency returns at 8-and 10-week horizons, concentrated in midsize, relatively volatile assets. Using a panel of 70 USDT-quoted tokens on Binance from January 2021 through March 2026, we show that a contrarian strategy of buying past losers and selling past winners, by forming Jegadeesh-Titman (1993) calendar-time overlapping portfolios, earns a 39.6% annualized return (Sharpe 0.96, Newey-West t = 2.10). Reversal is stronger among high-volatility assets, generating a Sharpe ratio of 1.37 (t = 3.19), and is strengthened outside of the largest assets, generating a Sharpe ratio of 1.69 (t = 3.80). The effect is robust across tercile, quintile, and decile sorts; skip period variants; inverse-volatility weighting; and temporal subsamples. A circular block bootstrap with 10,000 replications corroborates the high-volatility result nonparametrically with 95% of Sharpe ratios above 0.67, and the high-versus-low volatility gap positive in 94% of replications. Several economic mechanisms to rationalize these findings are discussed, including the tendency of treasury managers to sell into upswings, and Nagel's (2012) theory that reversal compensates liquidity providers.
This paper introduces Credance-Based Collateral Exchange (CBCE) as a formal category of financial instrument distinct from conventional repo and from existing distributed ledger collateral protocols. Building on the concept of credance-the collective anterior belief that makes a financial transaction possible before any track record exists (Laverriere, 2026)-we argue that a significant class of collateral exchange activity operates on the basis of credance rather than documentation. We formalise credance as a time-varying bilateral function C(Pâ, Pâ, t), define the credance threshold θ as a composite score of transaction history, temporal depth, and costly honouring, and propose a three-mode typology of CBCE instruments. We analyse the conditions under which each mode is optimal, examine the implications for Islamic repo market development and for distributed ledger technology design, and argue that blockchain-based collateral protocols have systematically failed to incorporate credance as a design variable. A credance-aware DLT architectureincluding a credance oracle, a zero-knowledge credance proof mechanism, and a variable collateralisation ratio governed by C(Pâ, Pâ, t)-would represent a genuinely novel class of financial instrument. The paper concludes with a research agenda including a call for collaboration between financial economists and DLT researchers to prototype the proposed architecture.
⢠DeFi stablecoin yields track FFR/SOFR, but with a distinct T+3 structural lag. ⢠A settlement-friction framework links fiat rails to the T+3 transmission lag. ⢠The lag is universal for both compliant USDC and offshore, unregulated USDT. ⢠Basis regressions reveal a predictable settlement wedge after policy moves. ⢠Robust tests rule out protocol outliers, macro trends, and weekend artifacts. Decentralized Finance (DeFi) stablecoin markets increasingly function as a shadow overnight dollar system, yet the speed at which U.S. monetary policy transmits to on-chain yields remains unclear. Focusing on the recent âHigh-for-Longâ regime (2023â2025), I study this pass-through using daily Aave V3 deposit rates for USDC and USDT. Guided by a simple conceptual framework of settlement frictions and arbitrage constraints, I estimate an ordered VAR that controls for equity- and crypto-market cycles. The results show that DeFi yields are tightly anchored to the Federal Funds Rate (and, in robustness, SOFR), challenging the âcrypto-decouplingâ narrative. However, transmission exhibits a distinct T+3 structural latency, universal across both compliant USDC and unregulated USDT, indicating an infrastructural, systemic friction rather than issuer-specific constraints. Robustness tests, alternative-explanations analysis, and quantity-based mechanism checks rule out protocol outliers, broader macro trends, and weekend artifacts, supporting an interpretation based on delayed settlement and execution across fiat rails. Complementary basis regressions provide a direct pricing implication: the on/off-chain spread exhibits a significant, predictable wedge during the settlement window that dissipates thereafter. The findings imply that despite algorithmic immediacy, DeFi remains constrained by fiat infrastructure, and that improving on-chain capital efficiency may require modernizing payment rails alongside issuer-focused regulation.
The pervasive volatility and structural complexity of decentralized assets present significant challenges for modern portfolio management. This paper introduces Coin Quest, a novel, high-fidelity cryptocurrency tracking and risk management platform designed to address critical shortcomings in existing market solutions, notably high data latency and the deficiency of robust quantitative risk tools. Our technical proposal mandates a resilient microservices architecture centered on Apache Kafka for high-throughput, low-latency data stream ingestion, ensuring real-time portfolio valuation across disparate exchanges and blockchains. The analytical core of Coin Quest implements the Monte Carlo Simulation (MCS) framework to compute Value at Risk (VaR) and the superior measure, Conditional Value at Risk (CVaR), recognizing the non-normal return distributions inherent to crypto assets. Furthermore, we detail specialized algorithms necessary for comprehensive tracking and valuation of complex Decentralized Finance (DeFi) positions, including the calculation of Impermanent Loss, and quantitative monitoring of NonFungible Tokens (NFTs) using floor price metrics. We conclude by outlining empirical validation requirements demonstrating the systemâs capacity to maintain sub-100ms data latency and confirming the superior predictive accuracy of the MCS-based risk model against traditional historical simulations in highly volatile market environments.
This study, among many others, provides an initial quantitative contribution to the emerging literature as well as existing empirical evidence regarding contagion risks across cryptocurrency markets over time. Using VAR (Vector Autoregressive) and SVAR (Structural Vector Autoregressive) models with Granger causality, along with Studentâs t-copulas, we find that Bitcoin is likely to act as an independent asset in this market, while Ripple and Litecoin tend to be recipients of contagion effects, and Ethereum appears to be a primary source of contagion. Our study offers additional insight into the investigation of contagion risks between both historical and future cryptocurrency values by employing Studentâs t-copulas for joint distribution analysis and aims to determine whether these contagion effects remain consistent over time. The results suggest, in both cases, that all cryptocurrencies tend to move negatively in extreme value conditions. Investors are therefore encouraged to pay closer attention to âbad newsâ and market movement patterns in order to make timely decisions regarding buying, holding, and selling. Note: This thesis reflects the state of cryptocurrency markets and related quantitative models as of 2019-2021. The findings and conclusions are intended to preserve the integrity of the research conducted during this specific time period. I acknowledges that this field evolves rapidly, and an updated analysis may be presented in a forthcoming research paper.
This review synthesizes the emerging literature on behavioral finance in cryptocurrency perpetual futures and perpetual swaps. It uses a constrained systematic review of accessible repositories, publisher pages, and citation trails for studies published or posted from January 2021 to April 2026. The synthesis separates 13 direct perpetual-futures studies from 6 adjacent behavioral studies that inform interpretation. The central question is how behavioral mechanisms shape trading, pricing, and market quality in perpetual futures markets. The strongest direct evidence concerns speculative demand and basis risk, leverage choice and liquidation risk, funding-rate carry and arbitrage behavior, informed trading and market quality, and exchange-design effects across centralized and decentralized venues. Direct evidence on classic behavioral constructs such as fear of missing out, overconfidence, disposition effects, and learning remains sparse in perpetual-specific settings. Three conclusions stand out. First, perpetuals are behaviorally distinctive because funding fees, leverage, mark-to-market margining, and auto-liquidation create a high-frequency feedback system between prices and trader positions. Second, the strongest causal evidence indicates that perpetual contracts increase spot-market trading volume but worsen adverse-selection conditions when informed trading rises around funding windows. Third, exchange design matters because inverse, linear, quanto, oracle-priced, and VAMM-based contracts expose traders to different incentives and liquidation dynamics. The most important research gaps concern trader-level identification, CEX-DEX comparisons using comparable data, contract-type heterogeneity, and causal tests of leverage-rule changes.
This thesis examines the implications of market frictions in international finance and macroeconomics in three contexts. The first chapter documents the effect of trading relationships on client trading outcomes in the over-the-counter (OTC) foreign exchange (FX) derivatives market. The second chapter documents the effect of nominal wage setting frictions on employment. The third chapter examines the behavior of non-U.S. central banks when firms engage in currency mismatch, borrowing more in dollars than given by their dollar operating exposures, emphasizing how imperfect regulation may affect U.S. dollar interest rates. In the first chapter, joint with Gerardo Ferrara, I study whether clients that rely more heavily on a dealer in the OTC FX derivatives market have worse trading outcomes after the dealer is adversely shocked. Using granular transaction-level data, we document that trading relationships are persistentâin an active trading week, clients are more likely to trade with a dealer that they had a relationship with and relied on more heavily. Then, we exploit the March 2023 collapse of Credit Suisse as an exogenous shock to exposed clientsâ set of trading alternatives when relationships are persistent. Using difference-in differences analyses, we find that, although Credit Suisseâs EURUSD notional traded and trade count declined, clients that relied less heavily on Credit Suisse did not differentially reduce their Credit Suisse-specific trading activity relative to more reliant clients. Instead, more reliant clients continued trading at the client level and increased activity with other existing dealer relationships without incurring additional costs, relative to less reliant clients. These findings suggest that search and bargaining frictions were not particularly costly for heavily reliant clients after the shockârelationship persistence did not differentially prevent them from reallocating activity to existing alternative dealers, or lead to relatively greater costs, when their relationship dealer came under stress. In the second chapter, joint with Gert Bijnens, Hugo Monnery, and Laura Nicolae, I empirically document the effect of wage changes, driven by wage indexation to inflation, on firm-level employment growth. In Belgium, nearly all employeesâ wages are indexed to inflation and firms are grouped into labor agreements that determine the exact timing and frequency at which wages are indexed, e.g. every year or every month. Using firm-level administrative data, we estimate two-stage least squares regressions of firm-level employment growth on wage growth, instrumented by the wage growth implied by the firmâs indexation policy. We find that employment contracts by 0.4% over four quarters for each 1% increase in wages. This result is robust to including NACE sector-date fixed effects and to using only variation in firmsâ indexation timing, controlling for their chosen indexation frequency. About one-third of the response comes via anticipation of future wage increases. The elasticity is more than twice as large in magnitude in the post-pandemic period than before it, suggesting strong nonlinearities. Overall, these results show that, by preventing inflation from reducing real wages, inflation indexation reduces employment. In the third chapter, joint with Mitali Das, Gita Gopinath, Taehoon Kim, and Jeremy Stein, I document an externality of central banksâ imperfect regulation of firms that engage in currency mismatch, which results from central banksâ dollar reserve accumulation decisions. We explore how foreign central banks behave when firms engage in currency mismatch. Using a panel of 56 countries, we document that central bank holdings of dollar reserves are correlated with the dollar-denominated bank borrowing of their non-financial corporate sectors. Then, we build a model in which the central bank can deal with private-sector mismatch, and the associated risk of a domestic financial crisis, by: (i) imposing ex ante financial regulations; or (ii) accumulating dollar reserves to serve as an ex post dollar lender of last resort. The model highlights a novel externality: individual central banks may over-accumulate dollar reserves, relative to what a global planner would choose. Under imperfect regulation of currency mismatch, individual central banks do not internalize that their hoarding of reserves exacerbates a global scarcity of dollar-denominated safe assets, which lowers dollar interest rates and encourages firms to further increase the currency mismatch of their liabilities. Relative to the decentralized outcome, a global planner may therefore prefer higher capital requirements and reduced holdings of dollar reserves.
This paper studies volatility prediction for Ethereum in the post-Merge era. Using daily ETH/USD returns from 15 September 2022 to 23 April 2026, we compare standard GARCH(1,1), Heston-Nandi GARCH(1,1), cross-validated and aggregated EWMA predictors, and Nadaraya-Watson kernelregression predictors. The kernel forecasts are constructed from a rank-transformed state vector that captures recent volatility and signed-return conditions, allowing the conditional variance function to be nonlinear and state dependent. The results show that forecast performance is strongly horizon dependent. At the one-day horizon, the kernel predictor using the fitted GARCH volatility state delivers the lowest final cumulative squared prediction error, outperforming the standard GARCH benchmark and all EWMA-type competitors. At the ten-day-ahead horizon, the advantage of local nonparametric information weakens, and the mean-reverting structure of GARCH becomes more valuable. The estimated kernel surface reveals that predicted ETH volatility is highest when elevated recent volatility coincides with negative signed-return pressure. Conditional quantile results further show that kernel-based VaR improves lower-tail risk forecasts, especially at the 1% quantile. Overall, the evidence suggests that post-Merge Ethereum volatility is persistent, asymmetric, heavytailed, and nonlinear, and is best modelled by combining economically meaningful volatility states with flexible nonparametric forecasting maps.
We use novel intraday data to study the price discovery process in cryptocurrency markets around U.S. monetary policy, inflation, and labor market announcements. Our analysis reveals the following: (1) volatility, trading volume, and bid-ask spreads rise sharply at announcement times and remain elevated for up to 30 minutes relative to comparable non-announcement intervals, indicating that cryptocurrency investors pay attention to these announcements and that information is quickly incorporated; (2) announcement surprises associated with higher yields or "risk-off" conditions cause substantial cryptocurrency price declines - pointing to a potential strengthening of U.S. monetary policy transmission to the real economy in recent years; (3) the time-varying magnitude and direction of price reactions more closely resemble those of U.S. equities than those of fiat currencies or commodities, suggesting that cryptocurrencies behave primarily as risk-sensitive assets; (4) price impact estimates of order flow around announcements are consistent with rising institutional participation in cryptocurrency markets and their crucial role for price discovery.
We construct a protocol-native valuation signal for Ethereum based on demand-side fees expressed as a share of token supply. The signal measures the log deviation of current fee intensity from its trailing median, a dimensionless ratio denominated entirely in ETH. It predicts subsequent token returns at 10 to 60 day horizons with in-sample R-squared up to 22.8% and expanding-window out-of-sample R-squared of 14.4% at 45 days. The signal retains predictive power after macroeconomic controls, standard crypto risk factors, and momentum controls, and predicts ETH-specific relative returns. Predictability emerges only after the Dencun hard fork (March 2024), which separated execution fees from data availability fees, making the demand signal empirically detectable. Our findings demonstrate that demand-side economic flows are capitalized into token prices in the absence of firms, contracts, or residual cash flow rights, extending valuation logic to rule-based economic systems.
This paper proposes a structured decentralized finance (DeFi) strategy designed to accumulate Ethereum (ETH) over time while exploiting market volatility through liquidity provision and controlled directional exposure. The framework combines concentrated liquidity provisioning on Uniswap v3 with a hedge position using low-leverage directional exposure and a reserve of stablecoins for counter-cyclical accumulation during market drawdowns. The strategy is implemented on Layer-2 networks-specifically Arbitrum and Base-to reduce transaction costs and capture diversified trading flows. We present a formal mathematical treatment of impermanent loss under concentrated liquidity, Monte Carlo simulations of ETH price paths under three market regimes, and an optimization framework for liquidity range selection. The proposed system transforms three distinct market conditions-sideways volatility, bullish breakouts, and market downturns-into opportunities for yield generation, directional gains, or asset accumulation. Results indicate that the hedged strategy achieves a superior risk-adjusted profile relative to unhedged liquidity provision across all tested volatility regimes.
This paper examines whether social media sentiment derived from Twitter and Reddit improves the explanation and prediction of cryptocurrency volatility. Using Bitcoin and Ethereum as benchmark assets, we combine sentiment indicators with GARCH-type models and the HAR-RV framework. Results suggest that cryptocurrency volatility is primarily driven by internal market dynamics rather than social media sentiment.
João Pires da Cruz, Daniel Costa, Pedro Granate, Armando Teixeira ¡ 6 authors
We study the formation and evolution of trading networks in non-fungible token (NFT) markets using transaction-level data from two major collections, Bored Ape Yacht Club (BAYC) and Azuki. We introduce a simple transaction-based clustering rule that identifies dynamically evolving trading networks formed by buyer-seller interactions. These networks correspond to persistent trading structures linking wallets through sequences of transactions. We document three main empirical regularities. First, trading networks emerge endogenously and exhibit heavy-tailed size distributions consistent with preferential attachment dynamics. Second, the internal connectivity of large networks displays scale-free degree distributions characteristic of growing trading systems. Third, the lifetime of trading networks follows approximately exponential statistics, indicating a memoryless extinction process. These findings suggest that NFT markets are organized around evolving clusters of trading relationships rather than isolated transactions. The results replicate across collections, indicating that trading network formation is a robust structural feature of NFT markets. Our findings provide new evidence on the microstructure of digital asset markets and the mechanisms governing the formation and persistence of trading relationships.
This paper investigates the extent to which cognitive heuristics, social influence, and digitally-mediated sentiment drive the extreme volatility of cryptocurrency, Decentralised Finance (DeFi), and Non-Fungible Token (NFT) markets, and the degree to which these dynamics deviate from the Efficient Market Hypothesis. Using an integrative narrative review and a synthesis of empirical evidence from 2014â2025, the paper develops the Integrated Digital Asset Behavioural Model (IDABM), a four-variable framework relating market stability to social velocity (Sv ), heuristic load (Hl ), platform gamma (PÎł ), and liquidity leverage (Ll ). The analysis draws on demographic and sentiment data, case evidence from the 2022 Terra/ Luna and FTX collapses, and a comparative cross-asset bias taxonomy. The findings indicate that digital asset markets constitute a pure sentiment environment in which the absence of conventional valuation anchors produces heuristic dominance and structurally amplified herding behaviour. The paper concludes that effective regulation must shift from informational disclosure toward behavioural guardrails â including algorithmic accountability, regulation of gamified trading interfaces, and behavioural literacy requirements.
We document a sizeable disposition effect in the market for non-fungible tokens (NFTs). Using a comprehensive transaction dataset from OpenSea, we show that NFT holders systematically realize gains prematurely while holding onto losses, mirroring behavior documented in traditional equity markets. Consistent with a high participation rate of retail investors and the lack of clear fundamental values, the effect is significantly more severe than in equity markets. We further find that the magnitude of the disposition effect attenuates in December, consistent with end-of-year tax-loss harvesting incentives, suggesting that on-chain transactions can be monitored by tax authorities. Finally, to address the NFT market's episodic illiquidity, we introduce a novel measure of the disposition effect based on the time-to-sale of listed assets. Our findings extend behavioral finance theory to digital-asset markets and provide new tools for studying the disposition effect in illiquid trading environments.