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 · 248 shown
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%
Large language models can extract richer signals from financial news than fixed sentiment lexicons, and recent work has explored feeding such signals into portfolio construction.
PAPER REPORTS · Sharpe 2.33, annualized return 95.9%, cumulative net 100.1%, max drawdown -18.3% — pure beta, GPT-4o mini, Student-t,… · Sharpe 2.44, annualized 100.0%, net 106.9%, max DD -18.3% — same configuration at 50 bps (2025)
This paper investigates whether textual tone derived from large language models (LLMs) can predict future stock returns. Using Korean news articles, we employ five LLMs to extract textual tones: BERT (KrFinBERT), DistilBERT, RoBERTa, ELECTRA, and Llama3.
PAPER REPORTS · BERT (KrFinBERT) equal-weighted quintile High-Low: 0.242% per day, t=3.772 (5.324% per month), May 2023-May 2024, gross… · BERT High-Low Fama-French 3-factor alpha 0.232% per day (t=3.582); 5-factor alpha 0.232% per day (t=3.566), same…
This study analyzes the microstructural mechanisms through which the rapidly expanding single-stock leveraged ETFs in the Korean capital market impede the price discovery function and amplify endogenous volatility.
Volatility is a fundamental characteristic of financial markets and plays a crucial role in investment decision-making, portfolio management, and financial risk assessment.
PAPER REPORTS · Out-of-sample RMSE 0.009552 (GARCH(1,1)) vs 0.009670 (EGARCH(1,1)), 230 one-step-ahead rolling forecasts, 30 May… · Out-of-sample MAE 0.007521 (GARCH) vs 0.007406 (EGARCH)
OUR BACKTEST · Sharpe 0.83 · Return +37.3% · Max DD -11.4%
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%
We study how differences in AI-generated financial recommendations are transmitted into individual portfolio choices.
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…
Thousands of SOFR derivatives are available in exchanges and OTC, but the market remains illiquid and incomplete.
The Triadic Stress Index (TSI) takes a network index whose four factors were first observed in soil microbiome co-occurrence networks and applies it, without alteration, to the correlation network of financial assets.
PAPER REPORTS · Out-of-sample F1@p90 = 0.447 (TSI with memory, 2016-2026, 897 windows, OFR 23-window crisis list; no transaction costs… · F1 gap vs Absorption Ratio = 0.273 (0.447 vs 0.174) out of sample 2016-2026, block-bootstrap 95% CI [0.095, 0.392],…
OUR BACKTEST · Sharpe 0.87 · Return +229.7% · Max DD -46.7%
A new class of software systems is transforming investment analysis. Large language model agents assembled into collaborative team structures including analysts, researchers, and risk managers are increasingly deployed across financial markets.
PAPER REPORTS · Author's own prior human practice (not the AI framework): 127.17% return vs 50.67% benchmark within sixteen months,… · Ranked first among 276 comparable peer funds nationally through the 2020 market-stress period (peer average negative);…
We calibrate credit default swaps and index tranches with elastically stopped Lévy processes: each firm defaults when the running supremum of a latent, spectrally positive distress process crosses an independent exponential barrier.
Crypto-listed equity perpetuals trade while the primary cash market is closed, yet still need a mark for margin, funding, and liquidation.
The paper considers the problem of variable selection for forecasting electricity spot prices.
PAPER REPORTS · BMT hourly rMAE averaged across six areas: 0.501 (2022-2025 out-of-sample, no transaction-cost concept; rMAE < 1 means… · BMT daily baseload rMAE_t averaged across six areas: 0.408 (2022-2025)
Cryptocurrency exchange-traded products (ETPs) listed on European exchanges provide a regulated environment for studying intraday market anomalies.
PAPER REPORTS · AUC-ROC up to 0.823 for one-bar-ahead DPOT prediction (LR, cumulative features, VIRBTC.ST), out-of-sample last 20% of… · AUC-ROC 0.821 for DPOT, LR, cumulative, VBTC.XE; 0.812 session-based
In this paper we investigate the information content of the lower part of the spectrum of financial correlation matrices, as a source of information on market synchronization.
OUR BACKTEST · CAPITAL EXHAUSTED · FAILED SANITY CHECK
How much capital a trading strategy can absorb before its edge disappears is a causal question about how much is deployed, but it is answered with observational proxies that rest on incompatible assumptions.
The daily return of a stock is often restricted to an exchange-imposed band to curb extreme fluctuations. Any attempted price movement beyond this band is clipped, leaving an unobserved excess.
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%
Completely monotonic inverse marginal (CMIM) utilities, introduced in [MSZ24], constitute a tractable class of preferences that includes many of the most important utility functions used in mathematical finance, such as power and exponential utilities.
OUR BACKTEST · Sharpe 0.48 · Return +15.8% · Max DD -20.6%
We consider a market maker who can only obtain and dispose of inventory by responding to a sequence of sealed-bid enquiries, and whose customers arrive with imbalanced intent: sellers more often than buyers, or the reverse.
Hawkes-based microstructural foundations for rough volatility, leverage, and rough Heston-type limits were developed by El Euch et al.
We develop a certified, scalable approximation for high-dimensional Wasserstein distributionally robust portfolio optimization. For expected-utility maximization under order-one Wasserstein ambiguity, standard duality yields a semi-infinite convex program.
PAPER REPORTS · ε=10^-2 (best DRO policy), 476-asset monthly rebalanced, 2021-2025, no transaction costs: cumulative return 3.64,… · ε=10^-4, 2021-2025, no costs: CR 4.77, σ 0.45, SR 0.93, MDD 0.38, Calmar 1.12
OUR BACKTEST · Sharpe 0.63 · Return +106.7% · Max DD -45.0%
We study Lambda-quantiles, a generalisation of classical quantiles in which the constant probability level $λ\in [0,1]$ is replaced by a functional parameter $Λ\colon \mathbb{R} \to [0,1]$.
OUR BACKTEST · Sharpe 0.68 · Return +249.4% · Max DD -84.8%
This paper documents an applied natural-language-processing framework for measuring the tone of Brazilian Monetary Policy Committee (Copom) statements.
Current portfolio construction methods are either agnostic to the effects of idiosyncratic shocks (standard factor models) or to the latent data structure driving systematic returns (recent graph-based approaches).
PAPER REPORTS · Contagion Cut (proposed): CAGR 21.0%, Sharpe 1.07, Calmar 0.611, Jan 2019-Mar 2026, 0 bps transaction costs · Contagion Cut: Sharpe 1.04 at 10 bps, 1.01 at 20 bps, 0.929 at 50 bps (CAGR 20.5%, 19.9%, 18.3%), Jan 2019-Mar 2026
OUR BACKTEST · Sharpe 0.75 · Return +143.1% · Max DD -40.3%
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%
Neural and numerical policy solvers can produce feasible controls even when the optimal rule and its binding constraints are unavailable.
The enormous growth in datasets, both in number and size, has prompted investors to adapt to new ways for assimilating information.
OUR BACKTEST · Sharpe 0.11 · Return +31.6% · Max DD -96.3%
The growth of decentralized finance (DeFi) and sustainability-linked investment markets has been rapid.
We study exponential-utility maximization for high-frequency trading in a discretized fractional Brownian motion model. Using spectral methods for stationary Gaussian sequences, we derive the asymptotic growth rate of the optimal certainty equivalent.
Risk-aware Q-learning (RaQL) provides a model-free, two-timescale estimator for dynamic risk objectives, but its finite-budget behavior remains fragile: fixed inner-loop hyperparameters can produce unstable value estimates, persistent Bellman residuals, and…
PAPER REPORTS · Scheme 6 out-of-sample (918 daily obs, chronological test set, after 5bp turnover costs, mean over 20 seeds): Sharpe… · Scheme 0 fixed-parameter baseline out-of-sample (same test set, after 5bp costs, 20 seeds): Sharpe 0.5628 (sd 0.2281),…
OUR BACKTEST · Sharpe 0.34 · Return +16.6% · Max DD -10.4%
Informed traders are supposed to need anonymity: they profit by hiding among the uninformed. A decentralized exchange now publishes the counterparty. Every committed order, cancellation, rejection, and fill carries a persistent pseudonymous wallet address.
PAPER REPORTS · One-second out-of-sample R2: 10.88% anonymous vs 12.31% with identity, +13.2% relative (t=9.2), ridge, evaluation July… · Gradient-boosted trees, one second: 19.48% -> 20.65%, +6.0% (t=5.0), same evaluation window, no costs
This paper studies the investment and insurance strategies of defined-contribution (DC) pension plans under the mean-variance framework. We consider a stochastic environment with time-varying interest rates, contributions, and mortality risk.
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%
The impact of web datasets on market prices has suggested the development of new sources of information, such as social media and web portals, indicating the possibility of an emergent phenomenon.
PAPER REPORTS · Out-of-sample one-step-ahead return forecast Mean Error improved ~10% (-0.05112 to -0.04284) with the open-information… · Out-of-sample RMSE improved ~0.01% (1.94575 to 1.94543) and MAE ~0.1% (1.52465 to 1.52311); no trading strategy,…
Cross-correlations between financial signals are neither scale-free nor amplitude-independent: they vary with the time scale over which they are measured and with the magnitude of the fluctuations that dominate the average.
PAPER REPORTS · Synthetic minimum-risk MMFC: 10-period 99% VaR 6.923... (stated as 5.923, sd 0.607) and 97.5% ES 6.048 (sd 0.594),… · In-sample empirical MMFC: lowest average monthly drawdown and lowest 10-day 97.5% ES among the five portfolios at every…
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 study seven major crypto-perpetual liquidation cascades (2022-2025), and in the largest of them we can watch the mechanism directly.
Fractional Brownian motion (fBm) exhibits attractive features for financial modeling, including long-range dependence, path roughness, and anomalous diffusion.
OUR BACKTEST · Sharpe 0.23 · Return +8.6% · Max DD -6.9%
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.
We introduce a framework for preference-robust decision making when preferences over risk are modelled through generalised distortion risk measures. Unlike distributional robustness, our approach addresses ambiguity in the risk functional itself.
OUR BACKTEST · Sharpe 0.27 · Return +30.1% · Max DD -23.4%
Automated market makers (AMMs) are typically interpreted and evaluated as decentralized exchanges.
PAPER REPORTS · VBIAX, monthly TE, Jan 2, 2014 – Jun 30, 2026: G3M Pareto-dominates (higher CAGR and lower TE) for gamma in [2.73%,… · EQL NAV, economic mandate, Jun 19, 2018 – May 29, 2026: G3M dominates for gamma in [3.22%, 7.09%]
OUR BACKTEST · Sharpe 0.71 · Return +133.0% · Max DD -42.0%
We develop a unified modeling framework that connects two distinct types of bubbles defined in the literature: the rational bubbles (aka P-bubbles), and the local martingale bubbles (aka Q-bubbles).
OUR BACKTEST · Sharpe 0.47 · Return +220.6% · Max DD -131.6%
We ask a representative sample to write prompts seeking spending and investing advice from LLMs, then simulate the lifetime effects of following the advice under realistic asset and labor market conditions.
We formulate an over-the-counter (OTC) market-making problem in which request-for-quote (RFQ) arrivals are modelled by general Hawkes kernels and fills are controlled thinnings of the exogenous request flow.
PAPER REPORTS · Exponential Hawkes benchmark objectives (Monte Carlo, 2x10^4 paths, T=1 day, no transaction costs modelled): benign —… · Near-critical regime objectives: exact HJB 48.80+/-0.24, Poisson 42.58+/-0.22, mean VR 48.37+/-0.25, noise-aware VR…
Agentic AI is gaining acceptance in asset management, but governance has not kept pace: 88% of surveyed finance professionals report no operational governance framework for agentic AI despite universal awareness of its deployment, and only 24 of 75 large U.S.
OUR BACKTEST · Sharpe -1.47 · Return -45.1% · Max DD -49.7%
This paper builds Path Portfolio Optimization: portfolio theory on a path-first framework in which the signature is the universal coordinate of the price path, and asks whether it survives estimation.
PAPER REPORTS · Cross-area lead-lag portfolio, sign-carrying excitation 1→2 with q=0.85: mean P&L +0.000189, s.e. · Cross-area, sign-carrying excitation 2→1 with q=0.85: mean P&L −0.000227, s.e. 0.000012, t=−18.92, annualized Sharpe…
OUR BACKTEST · Sharpe 0.45 · Return +21.4% · Max DD -19.5%