https://www.ml-quant.com/topics/trading-microstructure-execution/Order books, market making, execution, high-frequency data and trading signals. en https://www.ml-quant.com/papers/arxiv/2610.02917/https://www.ml-quant.com/papers/arxiv/2610.02917/Fri, 09 Oct 2026 07:00:00 +0000MBOFormer and MBOFusion, two causal transformers with under 70 kilobyte serialized state, predict limit order book price trends with sub-millisecond latency across three markets and four prediction horizons. https://www.ml-quant.com/papers/arxiv/2610.05740/https://www.ml-quant.com/papers/arxiv/2610.05740/Fri, 09 Oct 2026 07:00:00 +0000Analysis of high-frequency limit-order-book geometry finds market states form a continuous structure with a dominant latent coordinate capturing between 83.66% and 85.27% of covariance variation. https://www.ml-quant.com/papers/arxiv/2610.05077/https://www.ml-quant.com/papers/arxiv/2610.05077/Fri, 09 Oct 2026 07:00:00 +0000Varying execution realism from ideal fills to latency and impact reshuffles rankings of LLM and classical trading policies, showing how backtest conventions affect headline results. https://www.ml-quant.com/papers/arxiv/2610.08169/https://www.ml-quant.com/papers/arxiv/2610.08169/Fri, 09 Oct 2026 07:00:00 +0000A regime-switching Hawkes process models intraday electricity order flow across liquidity conditions, showing that stress mainly strengthens self-excitation rather than reshaping cross-side dynamics. https://www.ml-quant.com/papers/ssrn/7553278/https://www.ml-quant.com/papers/ssrn/7553278/Fri, 09 Oct 2026 07:00:00 +0000Applies a strict pre-registered protocol to eleven trading signals across equity and futures markets, finding only trend survives development and hold-out periods with borderline significance after multiple-testing adjustment. https://www.ml-quant.com/papers/ssrn/7551520/https://www.ml-quant.com/papers/ssrn/7551520/Fri, 09 Oct 2026 07:00:00 +0000Well-documented Indian equity anomalies replicate strongly in-sample but all fail to survive 12 to 32 basis point retail transaction costs; required directional accuracy reaches 72% at hourly horizons. https://www.ml-quant.com/papers/arxiv/2610.01115/https://www.ml-quant.com/papers/arxiv/2610.01115/Fri, 02 Oct 2026 07:00:00 +0000The research measures when a trading signal accumulates enough statistical evidence for deployment before its economic value decays, deriving exact feasibility thresholds. https://www.ml-quant.com/papers/arxiv/2609.33058/https://www.ml-quant.com/papers/arxiv/2609.33058/Fri, 02 Oct 2026 07:00:00 +0000A survey traces how market design has evolved from floor trading through electronic exchanges to AI agents, examining whether markets can allocate goods efficiently without full revelation. https://www.ml-quant.com/papers/arxiv/2609.38545/https://www.ml-quant.com/papers/arxiv/2609.38545/Fri, 02 Oct 2026 07:00:00 +0000The research shows when to follow or fade financial narratives by studying the covariance between institutional statements and revealed trading positions using machine learning. https://www.ml-quant.com/papers/arxiv/2609.31491/https://www.ml-quant.com/papers/arxiv/2609.31491/Fri, 02 Oct 2026 07:00:00 +0000A lightweight module adds confidence estimates to limit-order-book forecasters, raising directional F1 by 0.11-0.15 when predicting the most confident 10% of trades. https://www.ml-quant.com/papers/arxiv/2609.35867/https://www.ml-quant.com/papers/arxiv/2609.35867/Fri, 02 Oct 2026 07:00:00 +0000A method generates realistic limit order book messages guaranteed to be consistent with current market state, achieving 100% replayability and 2.7–3.6× speedup over existing approaches. https://www.ml-quant.com/papers/arxiv/2609.32848/https://www.ml-quant.com/papers/arxiv/2609.32848/Fri, 02 Oct 2026 07:00:00 +0000Measurement of over a year of CME market data reveals transactions cluster within microseconds, yielding design principles: single-thread receivers never queue, two-thread splits reduce tail latency. https://www.ml-quant.com/papers/arxiv/2610.01325/https://www.ml-quant.com/papers/arxiv/2610.01325/Fri, 02 Oct 2026 07:00:00 +0000Reinforcement learning agent combining policy optimization with regime priors controls drawdown while blending learned and rule-based portfolio exposure on held-out equity data. https://www.ml-quant.com/papers/arxiv/2609.31260/https://www.ml-quant.com/papers/arxiv/2609.31260/Fri, 02 Oct 2026 07:00:00 +0000The study shows that reinforcement-learning traders in limit order books create phase transitions between orderly trading and volatile cascades, with non-linear market impact dynamics. https://www.ml-quant.com/papers/ssrn/7520438/https://www.ml-quant.com/papers/ssrn/7520438/Fri, 02 Oct 2026 07:00:00 +0000Order flow imbalance predicts contemporaneous mid-price changes in Kalshi binary event contracts, with explanatory power varying from 0.29 for sports to 0.02 for macroeconomic events. https://www.ml-quant.com/papers/ssrn/7545019/https://www.ml-quant.com/papers/ssrn/7545019/Fri, 02 Oct 2026 07:00:00 +0000Correcting for timestamp mismatch between ETF and Bitcoin markets, the study finds that same-day flows predict next-day returns at 1.67 percentage points per billion of inflow, while prior Bitcoin returns predict flows. https://www.ml-quant.com/papers/ssrn/7526338/https://www.ml-quant.com/papers/ssrn/7526338/Fri, 02 Oct 2026 07:00:00 +0000Bitcoin and Ether trading activity on major venues increases during the UTC window aligned with New York market open, with the effect appearing on both assets. https://www.ml-quant.com/papers/ssrn/7530158/https://www.ml-quant.com/papers/ssrn/7530158/Fri, 02 Oct 2026 07:00:00 +0000A survey traces how market design rules evolved from floor trading to limit order books to automated market makers to agent-based trading systems. https://www.ml-quant.com/papers/ssrn/7540819/https://www.ml-quant.com/papers/ssrn/7540819/Fri, 02 Oct 2026 07:00:00 +0000On-chain prediction market analysis finds 68.2 percent of volume sits in well-calibrated markets with average seven-day pricing error of 0.03. https://www.ml-quant.com/papers/ssrn/7542098/https://www.ml-quant.com/papers/ssrn/7542098/Fri, 02 Oct 2026 07:00:00 +0000Multilevel order flow imbalance in Indian equity derivatives predicts mean-reverting price dynamics with information coefficients growing from -0.004 at ten seconds to -0.032 at sixty seconds. https://www.ml-quant.com/papers/repec/boe-boeewp-023631/https://www.ml-quant.com/papers/repec/boe-boeewp-023631/Fri, 02 Oct 2026 07:00:00 +0000Comparing FX forwards in identical currency pairs and maturities, the study finds large pricing variation across dealers reflecting clientele and pricing power rather than funding costs. https://www.ml-quant.com/papers/arxiv/2609.22785/https://www.ml-quant.com/papers/arxiv/2609.22785/Fri, 25 Sep 2026 07:00:00 +0000The research extends adversarial reinforcement learning for market making to handle self-exciting order arrivals and price impact, using an LSTM module to improve robustness in complex microstructure environments. https://www.ml-quant.com/papers/arxiv/2609.26606/https://www.ml-quant.com/papers/arxiv/2609.26606/Fri, 25 Sep 2026 07:00:00 +0000Develops a nested optimization model for market making and rebate design in option markets, showing how exchanges can set fees to incentivize liquidity provision and improve market depth. https://www.ml-quant.com/papers/arxiv/2609.27786/https://www.ml-quant.com/papers/arxiv/2609.27786/Fri, 25 Sep 2026 07:00:00 +0000Extends the Almgren-Chriss optimal execution framework to enforce intertemporal cash constraints, reducing peak cash drawdown while maintaining implementation shortfall in multi-asset rebalancing. https://www.ml-quant.com/papers/arxiv/2609.26861/https://www.ml-quant.com/papers/arxiv/2609.26861/Fri, 25 Sep 2026 07:00:00 +0000Experiments show that algorithm design features like warnings, pre-configured strategies, and LLM advice raise market prices by increasing starting prices and fostering cooperative algorithm designs. https://www.ml-quant.com/papers/ssrn/7500483/https://www.ml-quant.com/papers/ssrn/7500483/Fri, 25 Sep 2026 07:00:00 +0000Q-learning pricing agents in simulated duopolies reach supracompetitive outcomes with no communication, achieving collusion indices of 0.778 and 40% profit gains over competitive benchmarks. https://www.ml-quant.com/papers/ssrn/7512338/https://www.ml-quant.com/papers/ssrn/7512338/Fri, 25 Sep 2026 07:00:00 +0000Analysis of Kalshi's regulated Bitcoin and Ethereum futures reveals that 39-48% of notional trades are mechanical fixed-size orders that vanish when fees are charged, indicating costless artificial volume rather than legitimate trading. https://www.ml-quant.com/papers/ssrn/7201787/https://www.ml-quant.com/papers/ssrn/7201787/Fri, 25 Sep 2026 07:00:00 +0000Hungary's 2015 adoption of payment-versus-payment settlement reduced currency excess returns by ten basis points, demonstrating settlement risk is a priced friction limiting arbitrage. https://www.ml-quant.com/papers/ssrn/7479619/https://www.ml-quant.com/papers/ssrn/7479619/Fri, 25 Sep 2026 07:00:00 +0000Semantic and tonal shifts across sequential Federal Reserve communications generate significant intraday price movements and abnormal volume, revealing incomplete information absorption at initial announcement. https://www.ml-quant.com/papers/repec/zbw-safewp-343098/https://www.ml-quant.com/papers/repec/zbw-safewp-343098/Fri, 25 Sep 2026 07:00:00 +0000Using German sovereign bond repo data, the research shows hedge funds are price-elastic in cash markets but highly inelastic in repo, inheriting elasticity from their cash-market counterparties.