arXiv subjects

Adam Darmanin

Publications and source records attributed to Adam Darmanin.

3 recordsLinked to original sources

arXiv2026

Retail Trader's Ruin: An Anatomy of Popular Signal Failure

We test whether five widely promoted retail signal families - trend, oscillator, candlestick, volume, and calendar rules - deliver a positive, economically meaningful, net-of-cost, and survivable edge. Practical viability is the conjunction of three predeclared gates: statistical edge after multiplicity correction, economic viability after trading costs, and finite-bankroll survival under leverage. Exposure-matched benchmarks, stationary-bootstrap confidence intervals, hierarchical Benjamini-Yekutieli control, one-sided claim-exclusion tests, and equivalence tests distinguish positive evidence, statistically refuted materiality, and unresolved cases. Four of six candidates - oscillator, volume, calendar, and candlestick - are REFUTED, ruled out on statistical and/or economic materiality grounds; trend and a momentum calibration benchmark are INCONCLUSIVE, with confidence intervals too wide at this sample size to resolve the claim; none is SUPPORTED. Cross-sectional tests use point-in-time membership and delisting corrections. The momentum benchmark itself does not clear the statistical gate and is classified INCONCLUSIVE, not REFUTED - the critical validity signature that a genuinely uncertain positive control is never falsely falsified by this design. Under FINRA- and ESMA-anchored leverage and margin scenarios, survival is not the binding constraint for any tested family at the US headline scenario, though it becomes discriminating for trend and oscillator under the higher-leverage EU CFD scenario. The results reject specific promoted deployability claims where confidence bounds rule out the declared effect threshold and classify the remaining cases as unresolved rather than treating non-significance as proof.

q-fin.ST↗

arXiv2026

LV-ROVER-MLT: Low-Resource Maltese OCR by Synthetic Fine-Tuning and Multi-Stream Arbitration

Maltese has substantial text corpora and pretrained language models, but paragraph-scale OCR training data remains scarce; NOMOCRAT provides 57 verified annotated pages. LV-ROVER-MLT combines synthetic fine-tuning of Tesseract~5 with five complementary recognition streams and lexicon-gated word-level arbitration adapted to Maltese diacritics and hyphenation. In the DocEng~2026 Maltese OCR competition, the system placed first with held-out CER 0.0074; the next-ranked submission scored 0.0161 and NOMOCRAT scored 0.0163. The same approach produced a significant improvement over stock Tesseract on Luxembourgish, while the Hungarian result was inconclusive. A 36,803-pair Maltese OCR corpus constructed from EUR-Lex and Wikipedia provides an additional paragraph-level resource. Code, model weights, and corpus data are public.

cs.CL↗

arXiv2025

Language Model Guided Reinforcement Learning in Quantitative Trading

Algorithmic trading requires short-term tactical decisions consistent with long-term financial objectives. Reinforcement Learning (RL) has been applied to such problems, but adoption is limited by myopic behaviour and opaque policies. Large Language Models (LLMs) offer complementary strategic reasoning and multi-modal signal interpretation when guided by well-structured prompts. This paper proposes a hybrid framework in which LLMs generate high-level trading strategies to guide RL agents. We evaluate (i) the economic rationale of LLM-generated strategies through expert review, and (ii) the performance of LLM-guided agents against unguided RL baselines using Sharpe Ratio (SR) and Maximum Drawdown (MDD). Empirical results indicate that LLM guidance improves both return and risk metrics relative to standard RL.

cs.LG↗