Research
A continuously updated feed of research papers that pass our automated relevance screening for systematic trading — plus every paper we have published a review of, whatever it scored. Particular focus on alpha hypotheses that can be formalised and tested. The Radar also covers portfolio construction, market risk and execution where the research is directly relevant to systematic investment processes. Follow new entries by RSS.
15,697 papers screened · 250 on the radar · 39 shown
We study deep hedging in the context of dynamics risk measures, where sequential decisions are time-consistent.
PAPER REPORTS · Terminal hedging loss CVaR95% at 1-year maturity, 10,000-path test set with initial state perturbation, 0.1%… · Mean terminal P&L, same setup (alpha=95%): log -1.1143 (std 1.7095), static +0.2161 (std 2.6423)
This study investigates the forecasting performance of machine learning models and traditional econometric volatility models in predicting daily stock price volatility across selected Southern African Development Community (SADC) markets from 02 January 2015…
Foundation models promise accurate forecasts with little or no task-specific training, but whether they can replace models designed specifically for electricity price forecasting remains unclear.
PAPER REPORTS · Total BESS arbitrage profit over 2021-2025 (1,826 days, 1 MWh battery, 25 EUR round-trip operating cost deducted per… · Total profit Poland: unlimited-bid best 106,685 EUR (TabPFN-3) vs Oracle 123,439 EUR, 86.4% of perfect foresight; Spain…
Abstract This study examines the role of different social media sentiment dimensions in explaining stock market volatility in Pakistan.
Small-cap-inclusive equity universes contain recently listed and intermittently traded securities, so enforcing a common look-back discards a substantial fraction of the available information.
PAPER REPORTS · Annualized five-session volatility 11.17%, January 2000 - December 2025, net of all modeled execution costs… · Sharpe 0.814, same period, net of execution costs
We propose a neural calibration method to construct a recombining binomial tree directly from a set of given option prices.
This paper investigates the dynamic response of Shanghai crude oil futures (INE) to international benchmark price shocks and evaluates the evolution of market maturity from its inception to early 2025.
Classical option-hedging methods like Black-Scholes delta assume constant, free rebalancing, which real markets don't allow. Deep hedging trains a neural network to handle these frictions directly, and prior work reports strong results.
PAPER REPORTS · Whalley-Wilmott (paper's best strategy), test period Sep 2023-Dec 2024, 11,546 episodes, 5bp round-trip cost: mean… · Whalley-Wilmott cost saving vs BS delta: -$1.79 per episode, 95% CI [-2.21, -1.39], p < 0.0001 (test period, 5bp cost)
This paper studies European option pricing in a regime-switching Heston-Hull-White framework.
PAPER REPORTS · In-sample (train, 2 Jan-6 Aug 2024, 8,286 obs) DL-RS-HHW pricing error: RMSE 0.0072, MAE 0.0050 (option prices in yuan;… · Out-of-sample (test, 7 Aug-30 Sep 2024, 1,234 obs) DL-RS-HHW pricing error: RMSE 0.0179, MAE 0.0097
Abstract Conventional financial market forecasting models are challenged by the non-stationarity, the existence of regime changes, the presence of structural breaks, and the phenomena of volatility clustering in financial markets.
PAPER REPORTS · Directional accuracy 87.5% ± 2.1, NIFTY-50 daily, 2010-2025, walk-forward validation, no transaction cost assumption… · MSE 0.012 ± 0.003 (normalized), RMSE 0.110 ± 0.014 (121.8 index points), MAE 0.084 ± 0.011 (93.1 index points), MAPE…
Algorithmic trading now represents a market exceeding $20 billion, where even marginal gains in signal robustness can translate into economically significant returns.
PAPER REPORTS · Hybrid ensemble, OOS 2025 (252 trading days), 2.2bp per trade per leg: total return 51.26%, Sharpe 2.44, Sortino 5.35,… · Hybrid ensemble, OOS 2025: annualised CAPM alpha 0.423 (p = 0.011), beta 0.048, Probabilistic Sharpe Ratio 0.960
OUR BACKTEST · Sharpe 0.75 · Return +696.5% · Max DD -365.1%
This paper develops a unified framework for assessing systemic risk and identifying contagion channels in the global banking system using a Temporal Heterogeneous Multiplex Graph Neural Network.
PAPER REPORTS · MSE 0.0309 on one-quarter-ahead change in log(1+CDS), out-of-sample, N=336 bank-quarter forecasts (sample 1998-2025,… · MAE 0.1342, out-of-sample
PAPER REPORTS · Baseline HRP: annualised Sharpe 0.741, total return +417%, max DD -85%, 76 monthly rebalances 2020-02 to 2026-05,… · HRP-family variants span Sharpe 0.701-0.749 over the same window and cost assumption (best HRP_Dynamic_94 0.749,…
Implied volatility surfaces summarise the option market and are central to many financial applications.
PAPER REPORTS · Surface point-forecast RMSE aggregated over 30 horizons: 0.01262 vs persistence 0.01343, +6.09% gain; MAE gain +3.45%;… · h=1: RMSE 0.00619 vs persistence 0.00535 (-15.72%); MAE -34.49%
OUR BACKTEST · Sharpe -0.00 · Return -1.0% · Max DD -329.1%
Abstract A key puzzle in finance is why algorithmic traders with advanced neural models sometimes fail to beat simple traditional strategies, while in other cases they clearly outperform them.
PAPER REPORTS · Cluster 0 (most efficient), test 2025-2026, net of 0.1% one-way costs: DDPG annualised return 34.72%, cumulative… · Cluster 0 best classical: HRP Sharpe 1.591, Calmar 2.3562, max drawdown -13.07%, annualised return 30.80%
OUR BACKTEST · Sharpe 0.78 · Return +25.3% · Max DD -11.7%
Deep hedging is a data-driven approach to learn hedging strategies. It relies on synthetic price paths generator, as real market data is often limited for training.
OUR BACKTEST · Sharpe 1.78 · Return +5.9% · Max DD -0.6%
We introduce Deep-MKV-TS, a path-dependent McKean-Vlasov framework for financial scenario generation. The stochastic dynamics are chosen by matching selected path and volatility features of generated scenarios to those observed in the data.
PAPER REPORTS · Frozen ES drawdown-risk decision, 123 held-out sessions (January-June 2026): Deep-MKV-TS average exposure 1.99 +/- 0.06… · Conditional-forecast CRPS (x1000, lower better) on the 123-session chronological held-out test, four-seed mean +/- sd:…
We develop a scalable adjoint-to-control framework for continuous-time portfolio choice under smooth pointwise constraints.
OUR BACKTEST · Sharpe 0.19 · Return +24.7% · Max DD -66.8%
Using neural networks for stock return prediction typically requires choices about depth and hidden-layer width that are difficult to connect to financial interpretation.
PAPER REPORTS · HNN (marginal): pooled out-of-sample R2 0.509% (vs zero forecast), 1987-2016 · HNN (marginal): gross equal-weighted decile long-short spread 3.95% per month, 1987-2016
Generative models of limit orderbook (LOB) data have advanced rapidly, but their evaluation often focuses on stylised facts and selected market statistics.
Limit order book (LOB) simulators are most useful to practitioners when they combine realistic market dynamics, computationally efficient sampling, controllable scenario generation, and the ability to generalize beyond the instruments seen during…
Cryptocurrency time-series forecasting is a challenging task because market data usually exhibit high noise, strong volatility, non-stationarity, nonlinear dynamics, and long-range dependencies.
PAPER REPORTS · Bitcoin h=96: MSE 0.175 ± 0.009, MAE 0.314 ± 0.008 (5 seeds, most recent 20% of data as test) · Dogecoin h=96: MSE 0.413 ± 0.012, MAE 0.374 ± 0.010
This study focuses on developing an AI-supported prototype for multiperspective interest rate forecasting that combines classical econometric models with modern artificial intel-ligence methods.
Diffusion generative models have rapidly emerged as powerful tools for modeling complex financial data.
OUR BACKTEST · Sharpe 0.53 · Return +79.3% · Max DD -50.5%
Financial volatility is regime dependent, yet incorporating regime information into neural networks can also destabilize training. This paper asks where such information should enter a neural cross-sectional volatility forecasting model.
PAPER REPORTS · U.S. panel, 30 walk-forward windows Apr 2018 - Oct 2025, mean over 30 seeds: IC 0.5469 +/- 0.0012, ICIR 6.14 +/- 0.06,… · IC advantage over capacity-matched MLP-L: +0.0048 full sample (p<1e-4); +0.0207 top market-vol decile; +0.0322 COVID…
Financial forecasting models are typically developed in full precision, yet production deployment often requires low-precision inference to reduce memory and computational cost. Post-training quantization (PTQ) enables such deployment without retraining.
PAPER REPORTS · FP32 mean daily IC (cross-sectional Spearman, averaged over test dates, 2018-2025 walk-forward): TSMixer 0.490±0.043,… · Baseline mean daily IC over the same folds: pooled HAR 0.408, persistence (trailing 5-day realized vol) 0.315
OUR BACKTEST · Sharpe 0.41 · Return +41.7% · Max DD -37.2%
Specialist training beats generalist scale when forecasting financial statements. To our knowledge, no prior work jointly forecasts complete financial statements beyond one year, yet in a discounted-cash-flow valuation most firm value sits past that window.
PAPER REPORTS · Forma change-space R^2 0.289 on full test sample, test period 2010-2024, no transaction costs applicable (forecasting… · Forma per-horizon R^2: 39.0% at h=1 falling to 22.5% at h=20 (full sample); 38.0% at h=1 to 23.7% at h=20 (LLM sample)
OUR BACKTEST · Sharpe 0.57 · Return +17.5% · Max DD -10.5%
Our primary goal is to forecast and empirically examine the evolution of the implied volatility (IV) surface, with particular focus on the dates of scheduled meetings of the Federal Open Market Committee (FOMC).
PAPER REPORTS · h=1 out-of-sample RMSE, calls, ConvLSTM on SVI surface without dummy: 0.085 (sd 0.003) vs random walk 0.092; test year… · h=1 out-of-sample RMSE, puts, ConvLSTM on SVI surface without dummy: 0.077 (sd 0.003) vs random walk 0.084; test year…
Foundation models for time series forecasting demonstrate impressive zero-shot generalization but often underperform on specialized domains such as high-frequency finance.
PAPER REPORTS · Mean per-day correlation 0.3730 (GatedLinear+RF, 40 test days per stock, Dec 2024-Jan 2026 sample, transaction costs… · Pooled correlation 0.5972 and cross-day correlation 0.5631 (GatedLinear+RF, same 40-day test window)
OUR BACKTEST · Sharpe -0.42 · Return -0.6% · Max DD -0.8%
Predicting financial asset returns remains one of the most difficult challenges in empirical finance, driven by the low signal-to-noise ratio and the semi-strong form of market efficiency.
PAPER REPORTS · Sector LSTM (main 3-layer, k=10, 1995-2024): mean daily long-short return 0.100% before costs, Newey-West t = 7.81,… · Sector LSTM appendix figures (after 2bp per half turn): mean daily return 0.053%, t-statistic 4.106, annualized return…
OUR BACKTEST · Sharpe -0.03 · Return -9.6% · Max DD -140.3%
Control policies optimized in simulation can perform poorly in the real system when the parameters $x$ of the simulator are estimated from limited data but the resulting parameter uncertainty is not represented inside the simulation.
PAPER REPORTS · In-simulator spectral risk (x100, warm-up H=T), BS-VOL: RLM-trained policy 10.15 on RLM paths and 10.33 on SLM paths;… · In-simulator spectral risk (x100, warm-up H=T), HESTON-CORR: RLM policy 19.87 (SLM paths) / 19.93 (RLM paths) vs SLM…
OUR BACKTEST · Sharpe -0.18 · Return -66.9% · Max DD -81.1%
This paper develops a unified mathematical theory of implied, local, and learned volatility surfaces.
Financial sentiment classifiers are commonly evaluated against human labels, but strong linguistic performance does not necessarily imply economically useful return predictability. This study separates these questions through two experiments.
PAPER REPORTS · One-day mean rank IC, 2019 Benzinga S&P 100 sample: FinBERT 0.0143 (largest), Financial-RoBERTa 0.0141, Naive Bayes… · FinBERT one-day long–short: gross total return 12.96%, annualized Sharpe 1.11, max drawdown -6.37% (2019, gross — no…
OUR BACKTEST · Sharpe -1.16 · Return -61.4% · Max DD -64.2%
We present a novel application of Neural Networks with Local Converging Inputs (NNLCI) to improve the efficiency of existing numerical methods for pricing multi-asset options.
Local-stochastic volatility (LSV) combines vanilla marginals with richer smile dynamics, but calibration requires a slow, noisy and sequential McKean--Vlasov fixed point. We learn a projection-consistent operator for the calibration triple.
PAPER REPORTS · Calibration latency 0.60 ms/surface vs 98.5 ms particle baseline (paired, same hardware, synthetic held-out states) · Vanilla repricing RMSE 58.2 +/- 3.3 bps on 8 held-out surfaces, 2 seeds (spread across seeds, not a confidence…
OUR BACKTEST · Sharpe 0.53 · Return +5.6% · Max DD -2.0%
We propose an arbitrage-aware latent flow-matching framework for unconditional implied volatility surface generation.
OUR BACKTEST · Sharpe -0.06 · Return -1.2% · Max DD -5.1%
Implied volatility surface forecasting is essential for option valuation, hedging,and risk management, but remains difficult because future surfaces are stochastic while pricing inputs must satisfy static no-arbitrage shape restrictions.
The development of online banking has brought about an increase in fraudulent operations, which is a major problem for banks.
Heavy-tailed diffusion models replace Gaussian noise by a Gaussian variance mixture: denoising Levy probabilistic models (DLPM) take the mixing variables i.i.d. across coordinates, while Student-t EDM shares one mixing variable per sample.
OUR BACKTEST · Sharpe -0.33 · Return -3.3% · Max DD -5.8%