ELARVEN / Quantitative Research

Research on what survives contact with execution.

Independent empirical work on market data, execution and model robustness. We publish methods, evidence and failure modes. not trading promises.

Current scope: Market microstructure, provider transfer, execution-cost attribution and model validation.
Publication: No fixed editorial quota. A note is released only when the result is sufficiently documented.
Correspondence: contact@elarven.com

Position

A backtest is evidence to interrogate, not a result to market.

Selected research

Five notes / Versioned / Public

RN-001

Execution / Cross-asset

A controlled falsification study of a cross-asset ranking signal that looked statistically useful before market frictions were introduced.

RN-002

Data provenance / Spread

Recovering the provenance of a historical spread field, reproducing it from the original source, then testing what survives a provider change.

RN-003

Market data / Validation

A three-hour source offset could be corrected. The spread structure still did not transfer. This distinction matters.

RN-004

Methodology / Validation

A research protocol for freezing model evidence without silently turning already-seen observations into prospective proof.

RN-005

Execution intelligence

Why a prediction metric, an executable decision and realized PnL are different objects-and why missing fill data breaks the chain.

Research discipline

ELARVEN separates statistical evidence, execution evidence and prospective evidence rather than collapsing them into one performance number.

Windows, rules and relevant model state are fixed before a promotion decision is made.

Spread, slippage and latency are treated as part of executability, not as a footnote after model selection.

Negative results remain part of the research record when they explain why a promising hypothesis should not be promoted.

For research correspondence and technical discussion.