Quantitative Research / Systematic Trading
FindingStructure inRandomness.
Independent Quantitative Researcher exploring financial markets through statistics, systematic experimentation, and computational research.
02 — About
Research
Over Prediction.
Markets do not owe anyone a pattern. The work is to separate what is probably real from what is merely convenient — and to be honest about how little is ever certain.
This practice is built around hypotheses, measurement and scepticism: design the experiment, quantify the uncertainty, try to break the result, and only then decide whether it deserves capital.
[PLACEHOLDER] Short personal biography, background and current focus go here. Nothing about employers, education or track record has been assumed.
- 01
Statistical inference
Treat every result as an estimate with uncertainty, not a fact. Quantify before believing.
- 02
Evidence-based experimentation
State the hypothesis first. Record every variation. Let the data disagree.
- 03
Robustness & reproducibility
A finding that cannot be regenerated, perturbed and re-examined is an anecdote.
- 04
Market behaviour
Markets are adaptive systems of people and incentives. Mechanisms outlast patterns.
- 05
Risk management
Survival is a precondition for compounding. Size positions for what is unknown.
- 06
Systematic decision-making
Rules written in calm conditions beat discretion exercised under pressure.
03 — Research
Notes from the
laboratory.
Working papers, experiments and methodological notes. The entries below are demonstration content that show the format — they contain no verified results.
- Statistical Inference
- Strategy Robustness
- Market Microstructure
- Portfolio Construction
- Time Series Analysis
- Systematic Strategy Research
- 01
Multiple Testing in Strategy Search
Why the best backtest in a large search is biased upward, and how to reason about it before looking at a single equity curve.
Statistical Inference14 SEPT 2026Demo entry - 02
Path Dependence and Drawdown Uncertainty
A single historical path is one draw. Resampling return sequences gives a distribution of drawdowns rather than one number.
Strategy Robustness30 JUL 2026Demo entry - 03
Notes on Order Flow Imbalance
A reading-note style outline of how signed volume and queue dynamics are commonly modelled at short horizons.
Market Microstructure18 MAY 2026Demo entry - 04
Covariance Shrinkage for Allocation
Sample covariance matrices are noisy when assets are many and observations are few. Shrinkage trades a little bias for a lot of variance.
Portfolio Construction02 MAR 2026Demo entry
04 — Systems
Quant
Brain.
A research operating system, drawn as a loop rather than a line. Ideas flow in, become hypotheses, are tested, and — the important part — what happens to the results flows back to change what is read, believed and proposed next. Remembering failure is a feature.
Conceptual diagram. Stage names and feedback ideas only — no thresholds, results, instruments or implementation details are shown.
Swipe the diagram sideways · tap a stage
Feedback loops
Pipeline stages
Feedback loop · human-in-the-loop
Curated learning
Patterns that survive across experiments become candidate knowledge. A human promotes them into the verified knowledge base — it is curated on purpose, never auto-dumped.
Path · Reflect → Atomize
05 — Mathematics
Uncertainty, made visible.
Three small instruments. Every figure runs on synthetic, seeded data — they illustrate ideas, not results. Series are separated by brightness, opacity and line style only.
06 — Writing
Slow
thinking.
Research notes, mathematical concepts, reading notes, engineering articles and essays. Authored in Markdown / MDX under content/writing.
- 01
What a Backtest Cannot Tell You
A backtest is a description of the past under a set of assumptions. It is evidence, not a forecast.
Essay12 OCT 2026 · 1 MINDemo entry - 02
Reading Notes on Sample Size
Template for a paper reading note — claim, method, assumptions, and what would change my mind.
Reading Note21 AUG 2026 · 1 MINDemo entry - 03
Designing a Research Pipeline for Reproducibility
Versioned data, pinned environments and deterministic runs — the unglamorous parts that make results trustworthy.
Engineering09 JUN 2026 · 1 MINDemo entry - 04
The Central Limit Theorem, in Practice
Where the normal approximation helps, where heavy tails make it fail, and how to tell the difference.
Mathematical Concept11 APR 2026 · 1 MINDemo entry
07 — Contact
Let's Discuss
Ideas.
- GitHubgithub.com/your-handlePlaceholder
- LinkedInlinkedin.com/in/your-handlePlaceholder
- Emailyou@example.comPlaceholder
- OtherSSRN / Google Scholar / XPlaceholder