I am an independent researcher working on systematic-trading methodology, walk-forward evaluation, and the statistical structure of strategy populations. The work here is published under Daru Finance — a sole-author imprint, not a fund and not a sales platform.
Background
I am a candidate in Economic Sciences at UNIP. Most of what I know about quantitative finance came from outside the curriculum: years of building backtesters, running thousands of walk-forward experiments, and trying to falsify my own results harder than anyone else would.
In parallel, I work as a Quantitative Systems Consultant for a firm I cannot name (under NDA). The role is research-side: building filters, stress-testing strategy populations, evaluating signal robustness, and production-grade backtester engineering. That NDA is exclusive — I do not take any other consulting work in the quant domain, and Daru Finance exists strictly for academic publication.
What this site is for
Daru Finance is where the public-facing portion of my research lives. The aim is simple:
- Publish empirical work with real receipts — code, data aggregates, reproducible figures.
- Resist the temptation to oversell. Most strategies don’t work; that finding is the work.
- Surface methods that survive real out-of-sample tests, and show the regimes where they fail.
What you’ll find
The flagship article Edge is in the Process walks through my central empirical claim: across 30 asset/timeframe combinations spanning crypto majors, mid-caps, and three forex pairs — and across the ~2.77 million strategy-windows that constitute the detailed published aggregates for BTC, DOGE, BNB, and SOL — no individual strategy carries credible out-of-sample edge. The same pool, evaluated through Daru Finance’s proprietary robustness filter, produces portfolios that clear institutional-style hurdles in strong regimes and signals its own failure in weak ones.
The Research section hosts the SSRN paper landing and links to its reproducibility package. The Lab lists open implementations of the firm’s public models — random matrix theory, Higher Criticism + knockoffs, topological data analysis on strategy populations, tail-EVT cross-asset coupling.
Posture
None of this is investment advice. I publish so that other researchers can replicate, poke holes, and improve the methods. If something here contradicts a result you have, the right response is to email me.