Research
Every research paper we have written up, newest paper first. New arrivals are screened automatically for topical fit with systematic trading; the ones that clear it are read, written up, and — where the idea is testable on our data — backtested. 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.
16,915 papers screened · 588 on the radar
Entropic risk optimisation is a general framework for pricing and hedging financial derivatives in incomplete markets that can be used to decompose P&L into market and model risk contributions.
Factor mining seeks to discover signals from financial data that predict future asset returns and guide portfolio construction. Automated factor mining now spans genetic programming, reinforcement learning, generative models, and large language model agents.
Market-regime models typically assume a finite set of discrete latent states. We examine whether high frequency limit-order-book dynamics exhibit distinct regime separation or apparent regimes result from discretising an underlying continuum, analysing deep…
PAPER REPORTS · Intraday volatility forecast (not a trading strategy): pooled expanding-year walk-forward OOS R2 0.6821 / 0.7273 /… · Next-window RV direction hit rate: 61.2% (30s), 60.5% (45s), 59.5% (60s) from the continuous predictive block;…
Finding which variables cause which others in multivariate time series, and at what lags, is central to science and policy, yet existing methods force a choice between flexible confounder adjustment, data-driven lag selection, and inference that controls the…
PAPER REPORTS · ACLE-VAR (the paper's adaptive-lag procedure, no confounders): weighted long-short, SPY dollar-hedged, open-to-close. · ACLE-VAR by strategy (Ann. Return % / Sharpe), 2004–2020, no costs: Naive 4.64/1.15; Top 50% 8.54/1.31; Top 25%…
OUR BACKTEST · Sharpe -0.33 · Return -2.2% · Max DD -4.6%
We study when to buy a share that will later be sold optimally, when the price follows a geometric multi-skew Brownian motion whose skew levels model support and resistance.
OUR BACKTEST · Sharpe 0.21 · Return +40.7% · Max DD -78.0%
This paper develops a reproducible framework for predicting gold-price direction and converting the forecasts into a currency-hedged SHFE-COMEX spread-trading strategy.
PAPER REPORTS · Hedged (USD/CNH) walk-forward strategy: annual return 14.8%, volatility 7.1%, Sharpe 2.08, max drawdown 4.2%, win rate… · Unhedged version, same setup: annual return 11.2%, volatility 9.0%, Sharpe 1.24, max drawdown 12.7%, win rate 61.3%, 52…
Taiwan’s central role in global semiconductor manufacturing exposes Taiwan-related ETFs to technology concentration, geopolitical uncertainty, and supply-chain disruptions, resulting in return distributions characterized by heavy tails, volatility clustering,…
OUR BACKTEST · Sharpe 0.37 · Return +28.9% · Max DD -34.3%
We develop a price-based framework for stochastic portfolio theory in which trading strategies are generated from nominal price weights and evaluated relative to a price-weighted benchmark.
PAPER REPORTS · Cap-based p-diversity-weighted portfolios, terminal log relative wealth vs cap-weighted index on U^S (2002–2021,… · Cap-based p-diversity-weighted portfolios, terminal log relative wealth vs cap-weighted index on U^P (2002–2021, no…
OUR BACKTEST · Sharpe 0.38 · Return +39.6% · Max DD -47.5%
The enhanced index tracking (EIT) portfolio selection problem aims to construct a portfolio that is expected to outperform a benchmark index.
PAPER REPORTS · M-ER, known mean/covariance, short selling allowed, OOS Jan 2020-Dec 2024, monthly rebalancing, transaction costs not… · M-TSV, known mean/covariance, short selling allowed, same period, costs not stated: mean return 0.048939, SD 0.150532,…
OUR BACKTEST · Sharpe 0.70 · Return +62.4% · Max DD -32.5%
We determine the leading logarithmic asymptotics of the fixed-maturity right tail of the SABR model with $β\in(0,1)$ and absorption at zero, for every correlation $ρ\in(-1,1)$.
A cross-sectional signal is a forecast vector over $d$ assets at each date; demeaned and normalized, it is a point on a sphere. Its city is the direction of its time-averaged vector in a common target-aligned frame, a compressed summary.
OUR BACKTEST · Sharpe 0.68 · Return +75.9% · Max DD -48.2%
Financial LLM agents are often evaluated by comparing their end-to-end returns with those of a baseline and testing the paired difference against zero.
PAPER REPORTS · Full agent (A0) vs frozen-rule baseline, 723 earnings events, 2020-2024: gross deployment value +15.2 bps/event… · Same comparison net of a pre-stated 40 bps cost per nonzero event position: -1.2 bps/event (95% CI [-59.8, +54.3]),…
Flashblocks divide a block's priority gas auction into shorter sequential auctions that commit transaction order before the block is complete. We ask how this auction cadence affects bidding and competition among automated arbitrageurs, or searchers.
Abstract The fact that the solution of the Markowitz model is numerically unstable is recognized as one of the principal reasons for its poor out-of-sample results when applied in practice.
PAPER REPORTS · Out-of-sample Sharpe ratio of the μp=β (δ-stable) portfolio, daily data 2000–2020, rolling window, net of 10 bp ×… · Same setup, DAX: 0.0099 (w=120), 0.013 (w=180), 0.0206 (w=240), 0.0312 (w=480).
OUR BACKTEST · Sharpe 0.66 · Return +49.0% · Max DD -31.1%
Return predictability in financial markets is self-erasing: once an anomaly is discovered and capital flows toward it, its excess return decays.
PAPER REPORTS · Detector-conditioned strategy (Monte Carlo, synthetic data only): per-period Sharpe 0.224 at β=0.4, window n=100,… · Same strategy, window sweep (synthetic): peak per-period Sharpe 0.217 near n=100 at β=0.4.
OUR BACKTEST · Sharpe 0.33 · Return +13.5% · Max DD -10.2%
networks and Transformer models for financial volatility forecasting, with specific applications to the CBOE Volatility Index (VIX) and cryptocurrency markets (Bitcoin, Ethereum).
The complex and nonlinear behaviors of financial markets make it exceedingly challenging for average investors to evaluate the performance of a specific asset or make meaningful predictions about its future behavior.
PAPER REPORTS · Directional hit rate (MDA) of the LSTM forecaster on a random 80/20 holdout validation set (27,433 samples, 159 stocks,… · Diebold–Mariano statistics vs baselines on the same validation set: 3.42 (p = 0.001) vs persistence and 2.15 (p =…
Reinforcement learning (RL) is increasingly used for financial optimal-control problems when complex dynamics make analytical strategies difficult to obtain.
PAPER REPORTS · Proposed method (frozen analytical policy + residual PPO), execution cost halved from 0.0012 to 0.0006, σ_ξ=100, full… · Same method at a_true=0.0012: change 0.00019 [-0.00041,0.00079]; at a_true=0.0018: change 0.00019 [-0.00034,0.00071]…
We propose an interactive robo-advising framework that learns personalized risk preferences from scores provided by clients.
Deep reinforcement-learning policies for order execution can vary substantially across training seeds, so apparent architectural gains may reflect favourable training realisations rather than reproducible properties of the architecture.
PAPER REPORTS · Mean implementation shortfall, out-of-sample Jan 16–20 2023, frictionless replay with no commissions or market impact,… · Per-run CVaR-95 under the same setup: DDQL 110.8, MoE K=8 122.2, DDQL-cap8 104.1 bps.
Large language models are moving from producing trading signals to writing the code that executes them.
Abstract We investigate statistical tests for multivariate realized volatility models which are estimated using realized covariance matrices computed from intraperiod (high-frequency) financial data.
OUR BACKTEST · Sharpe 0.54 · Return +24.3% · Max DD -22.3%
Forecasting stock index prices remains a challenging task because financial markets are volatile, non-stationary, and influenced by several economic and behavioral factors.This study evaluates 12 supervised machine learning models against a Naïve Persistence…
Neural network-based approaches have emerged as efficient alternatives to traditional optimization-based procedures for the calibration of stochastic volatility models.
Short-horizon price-trend prediction from limit order books in equity and intraday electricity markets requires models that combine predictive quality with low single-sample latency and a small serialized model size to keep pace with rapid and continuous…
PAPER REPORTS · Classification performance only. These are not trading returns: no Sharpe, PnL or cost assumption is reported, and the… · MBOFormer test macro-F1 (mean of 3 seeds), h=10/20/50/100: - EPEX (3 test days, March 2021): 0.651/0.577/0.517/0.472 -…
In private wealth management, a manager delegating to artificial intelligence (AI) acts as the client's agent and the system's principal.
Abstract Estimating value-at-risk on time series data with potentially heteroskedastic dynamics is a highly challenging task.
PAPER REPORTS · LSTM VaR estimator, Fama-French 25 BM/OP portfolios, VaR 1%, n=250, test period ending 08.03.2019 (~500 backtest days):… · LSTM VaR estimator, Fama-French 25 portfolios, VaR 1%, n=250, test period ending 28.02.2022 incl.
We study whether model diversity survives selection into trading. In synthetic markets with a fixed mixture of three language-model families, news presentation changes their representation among submitted orders.
The increasing volatility of electricity prices driven by renewable energy integration, market shocks, and regulatory changes has reinforced the need for forecasting methods that go beyond point predictions and accurately describe the full conditional price…
PAPER REPORTS · JSU(4), the paper's preferred specification: average MAE 19.39 (SD 0.12) over the 24 hourly series. · JSU(4) average pinball loss: tau=0.99 1.29, tau=0.95 3.41, tau=0.05 3.53, tau=0.01 1.13. Same evaluation setup.
When developers change one component of an agent, such as its controller, a learned model or its verifier, they usually judge the change by an aggregate task score.
PAPER REPORTS · ARC-AGENT (full): Sharpe 0.56±0.29 (across-seed s.d.), MaxDD 10.5%, mean per-rebalance turnover 18.1%. · ARC-AGENT Sharpe excess vs static EW/MinVar frontier at matched drawdown: −0.04 (synthetic U3, 8 bps costs).
Reinforcement learning for trading often struggles to balance upside participation with drawdown control.
PAPER REPORTS · SPY, test 2020–2022, seed 42, 0.1% proportional transaction cost: total return 27.62%, annualized return 8.48%, Sharpe… · SPY, test 2020–2022, mean ± std over seeds {7,11,19,23,42}, 0.1% cost: total return 0.2725 ± 0.0109, Sharpe 0.6219 ±…
OUR BACKTEST · Sharpe 0.89 · Return +42.3% · Max DD -15.6%
In this work, we study the pricing of American options under stochastic local volatility (SLV) models extended by including stochastic correlation driven by an additional stochastic process.
In this paper we study whether a trading signal can accumulate enough statistical evidence for reliable deployment before its economic value decays.
PAPER REPORTS · Funding-rate lifetime-score signal, top quintile (109 events): 100/109 events with positive next-seven-day…
We study finite-horizon portfolio optimization with proportional transaction costs and trading opportunities arriving at the jump times of a Cox process.
OUR BACKTEST · Sharpe 0.52 · Return +0.0% · Max DD -0.0%
On April 20, 2020, the May front-month WTI oil futures contract, one day before its expiration date, opened near $\$17/$barrel and dropped far below zero in a single trading day, reaching an intraday low of $-\$40.32$ and settling at $-\$37.63$.
Liberalisation has been a catalyst for the strengthening of India’s banking sector; however, financial distress, fraud, and unethical practices continue to threaten the stability of the financial and stock markets.
In this paper, we extend earlier findings of supra-competitive outcomes in optimal-execution games by identifying a learned punitive mechanism that deters deviations and provides behavioural evidence of collusion.
We derive a short-maturity expansion for up-and-out put barrier option prices under continuous stochastic volatility when the strike and the barrier approach the spot at the diffusive scale.
We study the anatomy of commodity risk by distinguishing micro, market-level, and economy-wide sources.
Abstract This paper addresses the issue of time inconsistency in recursive stochastic control problems, in which the forward state process evolves under the influence of an additional recursive utility system.
Large language models are strong general-purpose code generators, but executable algorithmic trading remains a demanding specialization target: a model must translate a natural-language strategy specification into correct program logic for a specialized…
We develop a spouse-protected tontine in which a first death changes the household state but generates no pool transfer.
PAPER REPORTS · Out-of-sample learned policy, payment-seeking (λ=0.3): expected cumulative real payments 1860.4 (thousand AUD over 35… · Out-of-sample learned policy, intermediate (λ=1): expected cumulative real payments 1704.8, CVaR_0.95 terminal…
Decision-focused learning (DFL) trains predictors through downstream objectives, but a different loss need not provide an independent parameter-update direction.
PAPER REPORTS · Neural DFL vs MSE, N=100, K=20, 9 annual test folds 2016–2025: annualized TE 0.0276 vs 0.0331 (−16.5%), one-sided… · Same neural folds, K=20: worst-fold TE 0.0360 vs 0.0564 (−36.3%), max tracking drawdown 0.0306 vs 0.0362 (−15.5%),…
OUR BACKTEST · Sharpe 0.71 · Return +69.7% · Max DD -35.0%
European option smiles determine the risk-neutral marginal laws of an asset, but not their intertemporal coupling, which is decisive for many applications.
Bank treasury portfolios must balance yield, liquidity, and interest-rate risk across bonds of different maturities.
PAPER REPORTS · DP mean terminal wealth over 6 months, starting from 1 (simulated DNS-VAR chain, unconditional-mean initial curve):… · DP mean terminal wealth over 12 months: 1.1801 at c=0; 1.1166 at c=0.01.
Machine-learning signals built from financial text treat what institutions say, and what the media repeat, as evidence about value. But whoever shapes a narrative may be trading against it.
PAPER REPORTS · Linear model, raw voices + theory features: out-of-sample cross-sectional IC 0.171±0.018 (simulated main market, 5… · Gradient boosting, raw voices + theory features: IC 0.118±0.019 (simulated main market, 5 seeds; costs not stated)
We study the short-maturity implied volatility and the skew stickiness ratio for baskets of assets with continuous, possibly rough, stochastic volatility.
Global shocks - ranging from climate-related disasters and geopolitical conáicts to pandemics - transmit rapidly across borders and shape Önancial market co-movements.
The article examines the investment universe of the Russian financial market under conditions of restricted access to global capital markets.