8000 luiguiHerrera (Luigui Herrera) · GitHub
[go: up one dir, main page]

Skip to content
View luiguiHerrera's full-sized avatar

Block or report luiguiHerrera

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
luiguiHerrera/README.md

Luigui Herrera

Quantitative Finance Research · Portfolio Research · Econometrics · Systematic Strategies

MSc Economics, Finance & Computing Science
Quantitative Finance emphasis · Based in Spain

I work on quantitative finance, portfolio research, econometrics, market regimes and machine learning in finance.

Research Focus

  • Quantitative portfolio research
  • Dynamic asset allocation
  • Market regimes
  • Econometrics and statistical validation
  • Robust backtesting
  • Portfolio risk and fragility
  • Machine learning and deep reinforcement learning in finance

How I Work

Question → Hypothesis → Data → Assumptions → Benchmarks → Test → Robustness → Conclusion

I prefer to:

  • define the question before building the model;
  • make assumptions, constraints, costs and data boundaries explicit;
  • compare against credible benchmarks;
  • keep model selection separate from out-of-sample evaluation;
  • distinguish performance from statistical evidence;
  • test whether results survive different periods, seeds, assumptions and specifications;
  • look for what breaks the result, not only what confirms it;
  • preserve the path from inputs to conclusions;
  • report negative results when they matter;
  • avoid false precision;
  • keep the work reproducible and inspectable.

I am not interested in making a model win.

I am interested in knowing what survives the test and what does not.

Selected Research & Tools

A falsification-oriented evaluation of TD3 for cross-asset portfolio allocation.

I test whether the result survives matched benchmarks, transaction costs, cash assumptions, multiple training seeds, walk-forward evaluation and statistical controls.

The result is mixed: TD3 remains competitive in some settings, but statistical superiority over the full benchmark universe is not established.

→ Research page

An open-source, local-first Python tool for analyzing portfolio fragility beyond standard return and volatility metrics.

It looks at concentration, dependency, co-movement, drawdowns, historical replay and explicit hypothetical stress.

→ Project page

My public quantitative investment research platform.

It includes market-regime analysis, statistical market levels, portfolio diagnostics, research experiments and published market reports.

→ luiguiherrera.com

Interests

Quant Research · Portfolio Research · Systematic Investment Research · Quantitative Finance · Research Collaboration

Pinned Loading

  1. drl-portfolio-evaluation drl-portfolio-evaluation Public

    Falsification-oriented evaluation of TD3 portfolio allocation under realistic costs, matched benchmarks and statistical controls.

    Python

  2. portfolio-fragility-lab portfolio-fragility-lab Public

    Local-first Python tool for portfolio concentration, dependency, drawdown and explicit stress analysis.

    Python

  3. Luiguiherrera-web Luiguiherrera-web Public

    Quantitative investment research platform for market regimes, statistical levels, portfolio diagnostics and published research.

    TypeScript 1

0