Bastien Buchwalter, Jean-Michel Maeso, Vincent Milhau
We develop a reproducible three-step protocol to clean daily cryptocurrency data from CoinMarketCap, one of the most used data providers in academic research. The procedure targets three recurring anomalies that distort market-level indicators: (1) extreme market-cap spikes, (2) one-day and multiday dips in Bitcoin dominance, and (3) abnormal trading volumes. Using more than 28,000 cryptocurrencies from 2014 to 2024, we show that the method modifies only a small subset of data while improving the reliability of key market indicators. We do not adjust prices or returns, preserving actual trading conditions. As an application, we construct dynamic investable universes using cleaned data and realistic constraints based on market capitalization and volume. This exercise shows that cleaning and filtering jointly produce more reliable universes, reducing spurious extremes and making them suitable for empirical asset pricing research and portfolio construction.
Abstract Bitcoin's price history follows an approximate power law in time, with \((R^2 = 0.947)\) over 2011--2026. We show that replacing uniform calendar time with activity-warped time ---where time advances faster during high-activity periods---improves both in-sample fit and out-of-sample prediction. Two warping signals are evaluated: price volatility (absolute daily log-returns) and on-chain transaction volume (daily USD value transacted). Both benefit from a power transform \((w_t^\gamma)\) that reshapes the weight distribution: \((\gamma = 2.41)\) for volatility (amplifying large-move days) and \((\gamma = 0.56)\) for transaction volume (compressing extreme spikes). Transaction volume emerges as the stronger signal, achieving \((R^2 = 0.958)\) in-sample and winning 8 of 9 walk-forward splits (mean \((\Delta R^2 = +0.414)\)). Volatility wins 5 of 9 splits but requires no external data. Transaction volume selects \((\alpha = 0)\) (pure warped time), while volatility retains a calendar component (\((\alpha \approx 0.4)\)). Neither signal benefits from smoothing. Despite being nearly uncorrelated (\((r = -0.007)\)), combining the two signals does not improve out-of-sample performance---each captures complementary but individually sufficient information about Bitcoin's growth dynamics.
Recent developments in distributed ledger technology, artificial intelligence, and decision-making agents hold the promise of radically transforming market infrastructures. Indeed, the emergence of Autonomous Market Infrastructure (AMI)—an open, fully automated, and decentralized set of market-related functionalities—is widely anticipated. Such infrastructures, serving agents capable of fully autonomous behavior, would enable fully automated trading strategies. Moreover, as AMI-based solutions require minimal human intervention, they could be implemented at a fraction of existing costs. This should bolster competition and democratization, as AMI is accessible to everyone and establishes a level playing field.
The decentralized finance market exhibits extreme volatility and complex nonlinear dynamics that pose significant challenges for accurate price prediction and risk management. Traditional time series models, including Long Short-Term Memory networks and Transformer architectures, struggle with either computational inefficiency in capturing long-rangedependencies or inadequate context retention across extended sequences. This research investigates the application of Structured State Space Models, particularly the Mamba architecture with selective state spaces, for modeling temporal dependencies in DeFi markets. The proposed framework addresses the limitations of conventional approaches by leveraging SSMs' linear-time complexity while maintaining superior long-sequence modeling capabilities through context-aware selective mechanisms. Our methodology integrates SSM architectures with DeFispecific features including on-chain transaction volumes, liquidity metrics, and market microstructure indicators. Experimental validation across multiple cryptocurrency pairs demonstrates that SSM-based models achieve competitive performance compared to attentionbaseTransformers while offering substantial computational advantages. The results indicate that selective state space mechanisms enable effective capture of both short-term volatility patterns and long-horizon price trends in decentralized markets. This work contributes to the emerginintersection of advanced sequence modeling techniques and blockchain-based financial systems, providing insights for algorithmic trading strategies and risk assessment frameworks in the rapidly evolving DeFi ecosystem.
ABSTRACT This study provides a comprehensive systematic review and bibliometric analysis of 125 peer‐reviewed articles on bid‐ask spread estimators published between 1987 and 2025. Using the PRISMA framework, we map the intellectual evolution of the field, identifying a significant shift from foundational parametric models to data‐driven approaches. While early research focused on simple covariance‐based metrics, the field has recently been transformed by significant technical advances. Our network analysis identifies five major thematic clusters ranging from market dynamics and liquidity definitions to microstructure in high‐frequency and volatile environments. We highlight a critical research priority: utilizing high‐frequency data to validate low‐frequency models for reliable application in unobserved contexts, such as emerging markets and decentralized finance (DeFi). The findings underscore the enduring relevance of estimators in construction of long‐span historical series and noise‐adjusted liquidity measures. Future research must bridge existing methodological silos by integrating behavioral finance perspectives and advancing real‐time analytics for fragmented, high volatile global markets.
Abstract Automatic trading systems cope with the needs of put out emotional biases from the trading operation of public assets. These systems place orders based on a price model that forecasts the future price of an asset. Those systems, developed by edge funds and institutional investors, are not available to the public, and extensive research in this field is worth the effort. In this research, we developed a short-term price model based on a neural network and used it to forecast the near-future price direction. More in depth, we introduced the feature extraction process and parametric labeling strategy to build an ML ready dataset that includes more than 400 cryptocurrencies. The model is then validated by building a trading strategy on the two most capitalized cryptos at the time of writing: Bitcoin and Ethereum. The validation uses a trading simulation that spans six years of historical data for Bitcoin and Ethereum, including both retrospective (backtest) and prospective (forward test) evaluations. The results demonstrate that the neural network-based model exhibits a very good generalization to patterns found in historical data, enabling predictions in future data within the trading simulation. In addition, a comprehensive analysis of the importance of features was conducted to enhance the interpretability and performance of the model. Finally, we test our model in a simulated trading session; it shows that, with a simple buy-only strategy plus a stop loss, the trading system limits the draw dawn during bear markets.
Shiyu Zhang, Zining Wang, Jin Zheng, John Cartlidge
Systemic risk refers to the overall vulnerability arising from the high degree of interconnectedness and interdependence within the financial system. In the rapidly developing decentralized finance (DeFi) ecosystem, numerous studies have analyzed systemic risk through specific channels such as liquidity pressures, leverage mechanisms, smart contract risks, and historical risk events. However, these studies are mostly event-driven or focused on isolated risk channels, paying limited attention to the structural dimension of systemic risk. Overall, this study provides a unified quantitative framework for ecosystem-level analysis and continuous monitoring of systemic risk in DeFi. From a network-based perspective, this paper proposes the DeFi Correlation Fragility Indicator (CFI), constructed from time-varying correlation networks at the protocol category level. The CFI captures ecosystem-wide structural fragility associated with correlation concentration and increasing synchronicity. Furthermore, we define a Risk Contribution Score (RCS) to quantify the marginal contribution of different protocol types to overall systemic risk. By combining the CFI and RCS, the framework enables both the tracking of time-varying systemic risk and identification of structurally important functional modules in risk accumulation and amplification.
This research establishes a formal topological framework for managing non- stationary market assets in portfolios by synthesizing high-dimensional chaotic dy- namics with industrial quality control and cryptographic verification. We introduce the Hala Operator as a state-dependent regulator capable of inducing Successive Controlled Collapse (SCC)—a process that maps continuous chaotic flows onto discrete, stable fixed-point constellations. By utilizing Taguchi Design of Experiments (DoE) for off-market robustness and Zero-Knowledge SNARKs for execution privacy, we provide a mathematically rigorous solution to the "Newtonian Trap" of market unpredictability. Formal proofs of global stability, dimension collapse via divergence analysis, and the uniqueness of the discrete constellation are presented.
The association between cryptocurrency and sustainability is a complex and growing topic. Given that such linkage requires a continuous investigation, this empirical research, unlike the existing literature, explores if the volatility dynamics of digital assets are driven by the changes in sustainability uncertainty. In doing so, we use a recently developed ESG-based sustainability uncertainty index (ESGUI) and examine its effect on the volatility dynamics of Bitcoin and Ethereum ETFs. Employing the mixed data sampling (MIDAS) approach shows that ESGUI exerts a negative effect on the realized volatility of cryptocurrency markets. One possible explanation for this linkage is that as sustainability-related uncertainty rises, investors tend to adopt sustainability practices and initiatives. This shift towards sustainable practices can result in more consistent and foreseeable long-term economic conditions, thereby reducing the volatility of financial markets including the digital asset class. Our analysis offers key implications to cryptocurrency investors.
Aleksei Adadurov, S. Barseghyan, Anton Chtepine, Antero Eloranta · 6 authors
This paper examines the impact of reducing Ethereum slot time on decentralized exchange activity, with a focus on CEX-DEX arbitrage behavior. We develop a trading model where the agent's DEX transaction is not guaranteed to land, and the agent explicitly accounts for this execution risk when deciding whether to pursue arbitrage opportunities. We compare agent behavior under Ethereum's default 12-second slot time environment with a faster regime that offers 1-second subslot execution. The simulations, calibrated to Binance and Uniswap v3 data from July to September 2025, show that faster slot times increase arbitrage transaction count by 535% and trading volume by 203% on average. The increase in CEX-DEX arbitrage activity under 1-second subslots is driven by the reduction in variance of both successful and failed trade outcomes, increasing the risk-adjusted returns and making CEX-DEX arbitrage more appealing.
The 2026 cryptocurrency market cycle has witnessed the emergence of a novel asset class that defies traditional financial categorization: the AI-Integrated Meme Asset (AIMA). This report provides an exhaustive analysis of this phenomenon, utilizing the trajectory of Act I: The AI Prophecy ($ACT) as a primary case study. We posit that the convergence of large language models (LLMs) and decentralized community coordination has created a new "meta" for liquidity formation, characterized by the transition from static meme imagery to dynamic, agentic interaction. Central to this analysis are two theoretical frameworks proposed herein: the "Spring Effect," a market mechanics model describing the kinetic release of accumulated volatility following suppression events, and "Cognitive HODLing," a behavioral finance concept drawing on Social Identity Theory and Kahneman’s Prospect Theory to explain the rigidity of social consensus in the face of founder betrayal. Through a synthesis of on-chain data, behavioral analysis, and the philosophical frameworks of Vitalik Buterin and Satoshi Nakamoto, this report argues that $ACT represents the pioneer of a "Decentralized Agentic Economy," where value is derived not from revenue, but from the resilience of the human-AI social fabric.
We develop an agent-based model in which inflation emerges from decentralized price-setting and credit-financed production in an endogenous-money economy. Firms operate under working-capital constraints, form market-based price expectations through heterogeneous adaptive learning, and set prices via cost-plus rules with endogenous mark-ups. Bank lending simultaneously creates deposits, while heterogeneous lending rates and credit rationing shape firms' financing costs and, through unit costs, their pricing decisions. The economy features interacting production and credit networks: intermediate-input linkages propagate cost shocks across supply chains, while bank--firm relationships transmit financial conditions across firms. The interaction of network-based pass-through, state-dependent pricing incentives, and evolving credit conditions generates inflationary regimes, including episodes driven by pricing cascades and feedback loops.
Spot trading on decentralized exchanges (DEXs) remains materially inferior to centralized exchanges (CEXs) in throughput, latency, and market-maker tooling, ceding global spot liquidity to opaque, non-custodial intermediaries. We present Vela, a spot exchange engine designed from first principles to recover CEX-grade performance while preserving the verifiability and self-custody properties of a DEX. The core engine is an optimized Rust state machine running entirely in memory, achieving a median per-operation latency of 1.08 microseconds (p50) — 4.7 times faster than the prior state of the art — and 57,300 operations per second under a realistic mixed market-making simulation across ten simultaneous markets. Exchange state is maintained in a Merkle Patricia Trie whose root is periodically committed to an underlying blockchain, anchoring state integrity to an external consensus mechanism. Verifiability is achieved through an optimistic zero-knowledge proving scheme: state updates are assumed valid by default, with a seven-day challenge window during which any party may submit a proof of incorrect execution, and an on-demand fast-finality path for users requiring immediate settlement. We introduce two features novel to DEX design: (1) a market-maker credit system enabling capital-efficient cross-market quoting analogous to CEX credit lines, implemented natively within the matching engine's state transition function with atomic collateral enforcement; and (2) private L3 market data feeds authenticated via server-issued nonce challenges and wallet signatures, substantially reducing market-maker exposure to adverse selection. We describe the full architecture, five performance optimizations including a Delta elimination that reduces p99.9 tail latency by 73%, flamegraph profiling findings, and decentralization mechanisms including forced inclusion via a delayed inbox. The Vela engine is released as open-source software under the MIT license at github.com/arpjw/vela.
The rapid expansion of the digital economy has exposed significant limitations in traditional economic frameworks, which struggle to explain phenomena such as algorithmic decisionmaking, data-driven value creation, and platform-based concentration. Existing approachesranging from production function extensions to platform models-remain fragmented and lack a unified micro-foundation. This paper proposes a behavior-centered framework to characterize economic forms and introduces the concept of economic morphology defined along four dimensions: agent structure, factor composition, behavioral pathways, and spatial distribution. Building on this framework, we define the Information Process Ratio (IPR) as a measurable indicator capturing the proportion of information-processing activities within economic behavior. Using IPR as a discriminant variable, we identify four major economic forms in human historyagricultural (IPR 10-20%), industrial (30-40%), service (50-60%), and digital (75-90%+). We show that the digital economy represents a distinct morphology, not a continuation of the industrial paradigm. Contemporary financial technology (FinTech) systems-high-frequency trading (HFT), decentralized finance (DeFi), and automated market makers (AMMs)represent extreme high-IPR regimes (95-99%), making them natural laboratories for testing the framework's predictions. We operationalize IPR using transaction-level proxies such as order-to-trade ratios (OTR), cancellation rates, and algorithmic trading share, enabling empirical application in financial markets. The framework generates testable implications linking IPR to transaction intensity, market concentration, returns to scale, algorithmic mediation, and high-frequency volatility. We further introduce the concept of IPR arbitrage, whereby economic activity flows toward higher-IPR systems, and propose a Financial Tension Index (FTI) to capture systemic strain in high-IPR environments. By shifting the analytical focus from agents to behaviors, this paper provides a unifying perspective for understanding the structural transformation of the digital economy and offers concrete implications for financial technology regulation, algorithmic market design, and systemic-risk monitoring.
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 markets—including 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 features—including rolling statistics, exponential moving averages, and lag/difference features—suggesting 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.
This paper proposes Global Adaptive Equity Pricing (GAEP), a novel AI-driven framework for moderating economic inequality through real-time, consumption-event-based price personalization. At each domestic purchase, biometric verification links to encrypted networth data to compute a progressive adjusted price using the Wealth Elasticity Pricing Equation (WEPE). Excess payments from higher-net-worth individuals fund a transparent Gini Moderation Fund (GMF) for AI-optimized redistribution targeting a blended Gini coefficient of ≈0.30. Tunable parameters enable governments to control moderation velocity, balancing equity gains against capital retention risks in wealth-attracting jurisdictions. Calibrated to Singapore's 2025-2026 data (income Gini after transfers and taxes: 0.379; market income Gini before transfers: 0.452; wealth Gini: 0.55; top 1% hold ~14%, top 5% ~33% of household wealth), agent-based simulations project 15-41% Gini reductions over 20 quarterly cycles. Ethical safeguards include zero-knowledge proofs, blockchain-audited aggregates (no personal data exposure), fairness audits, and positive incentives. GAEP extends Gini theory and computational economics by integrating biometric technology with redistributive algorithms, distinct from usage-tiered tariffs or surveillance pricing. It offers policymakers a pathway for dynamic, consumption-led equity in AI-augmented economies while preserving innovation incentives.
Generative and agentic artificial intelligence is entering financial markets faster than existing governance can adapt. Current modelrisk frameworks assume static, well-specified algorithms and onetime validations; large language models and multi-agent trading systems violate those assumptions by learning continuously, exchanging latent signals, and exhibiting emergent behavior. Drawing on complex adaptive systems theory, we model these technologies as decentralized ensembles whose risks propagate along multiple timescales. We then propose a modular governance architecture. The framework decomposes oversight into four layers of "regulatory blocks": (i) self-regulation modules embedded beside each model, (ii) firm-level governance blocks that aggregate local telemetry and enforce policy, (iii) regulator-hosted agents that monitor sector-wide indicators for collusive or destabilizing patterns, and (iv) independent audit blocks that supply third-party assurance. Eight design strategies enable the blocks to evolve as fast as the models they police. A case study on emergent spoofing in multiagent trading shows how the layered controls quarantine harmful behavior in real time while preserving innovation. The architecture remains compatible with today's model-risk rules yet closes critical observability and control gaps, providing a practical path toward resilient, adaptive AI governance in financial systems.
We study the epistemic efficiency of decentralized prediction markets under autonomous agentic liquidity. We introduce the information-incentive gap (G) – the discrepancy between ground truth and the market-implied probability – and establish, via Itô's calculus and exact solution of the resulting moment ODE, exponential convergence of its second moment together with an explicit upper bound for the gap of order O(σ/λ−−√). A two-level empirical study on information-driven event categories (Politics, Economics, Finance, Crypto Markets), drawing on approximately 40 million time-series records collected over the study period, is consistent with the model: (i) platform-level analysis of N=100 resolved binary events per platform shows the mean gap decreasing from G¯=0.517 at T−168 h to G¯=0.229 at T−30 min for Kalshi, and from 0.583 to 0.002 for Polymarket, with an empirical convergence rate λemp≈1.4×10−6 s−1; (ii) a paired cross-platform comparison of N=34 matched event groups shows that Polymarket exhibits a lower mean gap than Kalshi (mean ΔG=0.27 at T−6 h; Polymarket leads in 85% of pairs), consistent with the theoretical dependence of convergence speed on liquidity-driven λ. Monte Carlo simulation (N=50000 paths) confirms a >276× reduction in convergence latency and a 109× improvement in the Information Efficiency Ratio (IER) compared to the human-centric baseline.
Cryptocurrency markets exhibit complex microstructural dynamics characterized by high-frequency volatility bursts, rapid regime switching, and long-range temporal dependencies, which expose several limitations of existing volatility forecasting approaches. In particular, attention-based models suffer from prohibitive quadratic computational cost on long high-frequency sequences, while many recurrent architectures struggle to adapt to regime transitions, asymmetric volatility responses, and risk-aware uncertainty estimation. To address these gaps, this paper proposesCryptoMamba-SSM, a novel volatility prediction framework built upon Mamba-based state space models with linear computational complexity. CryptoMamba-SSM integrates selective memory mechanisms with structured state space representations to effectively capture critical market microstructure signals arising from liquidity shocks and sentiment transitions, while dynamically adjusting memory retention across different volatility regimes. This design enables efficient modeling of long-sequence dependencies inherent in cryptocurrency price movements without incurring the computational bottlenecks of traditional attention-based architectures. Through comprehensive experiments on Bitcoin historical data spanning multiple market regimes, we demonstrate that CryptoMamba-SSM consistently outperforms conventional LSTM, GRU, and Transformer baselines, achieving up to a 23.7% reduction in Mean Absolute Error and a 31.2% improvement in directional accuracy. The selective memory mechanism effectively captures regime-switching behaviors and microstructural anomalies, leading to more reliable short-term volatility risk quantification. Moreover, the linear-time complexity of CryptoMamba-SSM enables real-time processing of high-frequency trading data while maintaining strong generalization across diverse market conditions.
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.
João Pires da Cruz, Daniel Costa, Armando Teixeira, João B. Duarte · 6 authors
We analyze the Ethereum transaction network using a spectral decomposition based on functional edge modes. Each transaction is represented as a complex amplitude indexed by the combined connectivity of the interacting addresses, and amplitudes are aggregated into mode-resolved coherent sums. Applying this construction to a snapshot of native ETH transfers from the second half of 2015 (∼1.9 × 10 6 transactions across 26,937 addresses), we identify spectral modes whose coherent power significantly exceeds that obtained under randomized phase baselines. Statistical significance is assessed via B = 1000 phase permutations with multiple-testing correction: 469 of 928 modes (50.5%) survive Benjamini-Hochberg control at the 5% false discovery rate, while none survive the more conservative Bonferroni threshold. Strong global coherence is primarily driven by high-degree nodes: removing the top 0.1% of nodes by degree (27 hubs) collapses the bulk of the spectrum towards the randomized baseline. However, statistically significant residual coherence persists across roughly half of the tested modes, indicating that organization in the network is not purely an artifact of hub aggregation. We frame these findings through a quantum-like analogy in which phase-aligned edge contributions interfere constructively, and discuss implications for the structural analysis of decentralized financial systems.