Empirical power-law frontier between predictive loss and structural forward work in LOB models extrapolates to held-out high-compute architectures with R²=0.941, motivating FastBiNLOB which exceeds SOTA macro-F1 at lower latency.
Kolm, Jeremy Turiel, and Nicholas Westray
5 Pith papers cite this work, alongside 28 external citations. Polarity classification is still indexing.
5
Pith papers citing it
28
external citations · Crossref
representative citing papers
Signature-linear trading rules reduce path-dependent statistical-arbitrage execution to one concave quadratic programme; fitted rules beat a z-score benchmark (9 vs 6 bps synthetic; 9 vs 2 bps on one pair).
A deep network trained jointly across S&P 500 stocks, using order-book predictors and volume commonality, roughly doubles out-of-sample predictability of intraday volume versus the CMEM benchmark.
A residual-learning hybrid of VAR and a feedforward neural network is applied to predict order flow imbalance, with reported gains over standalone models on Binance data.
citing papers explorer
-
Signature-Based Optimal Execution for Statistical Arbitrage with Path-Dependent Trading Signals
Signature-linear trading rules reduce path-dependent statistical-arbitrage execution to one concave quadratic programme; fitted rules beat a z-score benchmark (9 vs 6 bps synthetic; 9 vs 2 bps on one pair).
-
Forecasting Intraday Volume in Equity Markets with Machine Learning
A deep network trained jointly across S&P 500 stocks, using order-book predictors and volume commonality, roughly doubles out-of-sample predictability of intraday volume versus the CMEM benchmark.