Skip to content

Once a personal learning project, Helios has become an educational mission. Breaking the knowledge gap. To democratize professional investing.

Notifications You must be signed in to change notification settings

axeljuan02/Helios-Engine

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

85 Commits
 
 
 
 
 
 

Repository files navigation

HELIOS — Democratizing professional investing.

HELIOS is the fastest evolving project at the crossroads of quantitative finance, programming, and investment education.
It began as a deep dive into the institutional investment world to understand how real portfolios are built, how risk is managed, and how professional models work in practice.

As I sought to relearn, better understand, and go beyond the finance concepts studied at ESSEC to apply them to my own portfolio, I faced a massive gap separating institutions from everyday investors.

Many modern platforms and “finance influencers” promote an illusion of easy, accessible investing, while the reality is far more complex.

This project was born from the belief that everyone should have access to the same elite models and tools once reserved for institutional desks. The ambition quickly grew: to build a professional-grade portfolio optimization engine that is powerful, transparent, and accessible, turning the complex theories and tools of Wall Street into something anyone can learn from and use.

Helios is no longer just an optimizer, it’s a mission to educate as much as it optimizes, demystifying institutional models, breaking down the gap, and empowering individuals to make truly informed investment decisions.

mvp.compressed.mp4

✅ Key Features

Helios is a multi-page Streamlit MVP that operationalizes a Python quantitative engine for portfolio construction and risk analysis. It is designed as an auditable, reproducible demonstrator suitable for academic review and as a foundation for a future SaaS.

Page Purpose Key outputs
Portfolio Setup Define tickers, dates, and weights; validate inputs and constraints Cleaned universe, normalized weights, session state
Efficient Frontier Compute constrained mean–variance frontier and optimal points Frontier curve, max-Sharpe and min-volatility portfolios, allocation tables
Monte Carlo Risk Simulate paths from estimated moments and correlations VaR and CVaR from simulated P&L, loss distributions, stress scenarios
Factor Attribution (FF5) Regress excess returns on Mkt–RF, SMB, HML, RMW, CMA Alpha, betas, t-statistics, R² and adjusted R², factor loading charts
Explanations Translate results into plain-English narratives Contextual summaries, decision notes, caveats
Export Persist selected figures and tables PNG charts, CSV summaries, reproducibility notes

Methods and metrics : Efficient Frontier (Markowitz), Monte Carlo simulation, and Fama–French 5 attribution with KPIs including CAGR, annualized volatility, Sharpe, Sortino, and Max Drawdown. Long-only constraints with per-asset caps, frequency-consistent risk-free handling, documented annualization, and reproducible seeds. Live data via Yahoo Finance and Fama–French, with the option to run on bundled sample datasets.


⚙️ Processing Pipeline

sora pipeline

🛤️ Roadmap

Module Status
Multi-strategy optimizer ✅ Done
Efficient Frontier ✅ Done
Monte Carlo VaR & CVaR ✅ Done
A mission, a vision, a new design ✅ Done
Fama-French model ✅ Done
Codebase refactor into a modular engine ✅ Done
AI/NLP integration ✅ Done
MVP WebApp development (Streamlit) ✅ Done
MVP Release & Beta testing ✅ Done
First Website development (Showcase + Link WebApp) 🚧 In Progress
WebApp Upgrade (Supabase for DB + Auth) Planned
GARCH Model & MC Upgrade Planned
Macroeconomics Agregator Model Planned
Expansion (FastAPI, React, webdev...) Planned

📁 About This Repository

This repository contains the Helios Core Engine, the quantitative foundation powering the Helios platform.

Initially, it was built simply to code financial models in Python and demonstrate to universities and professionals both my understanding of financial principles and my ability to implement them programmatically.
It is now part of a broader project that includes a multi-page Streamlit MVP, with a SaaS-oriented WebApp in private development.

Detailed explanations of each component are provided in their dedicated documentation:


👤 Founder & Vision

Axel JUAN
BBA Third-Year Student @ ESSEC Business School
CFA Level 1 Candidate
Personal portfolio manager since age 15

Helios is part of a broader academic and entrepreneurial journey, combining technical rigor with a long-term vision: to give investors the knowledge and tools once reserved for professionals.

It is under active development and is designed to showcase the engine’s capabilities, coding standards, and quantitative rigor for academic admissions and professional evaluation.

For demo access, collaboration inquiries, or strategic discussions: [juan.axel@protonmail.com]

About

Once a personal learning project, Helios has become an educational mission. Breaking the knowledge gap. To democratize professional investing.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published