Founder & CEO · Alpha Stochastic Research
Quant Research Trainee · D+A Strategies
Engineering Student · Université de Technologie de Troyes
Advancing quantitative finance through mathematics, statistics, stochastic modelling, machine learning and open science.
I am an engineering student at the Université de Technologie de Troyes (UTT) and the founder of Alpha Stochastic Research, an independent quantitative finance research laboratory.
My work focuses on the intersection of quantitative finance, financial mathematics, stochastic processes, statistics, machine learning, and scientific computing.
My objective is to develop rigorous and reproducible quantitative methodologies that connect mathematical theory with real-world financial decision-making.
| Organization | Role | Focus |
|---|---|---|
| Alpha Stochastic Research | Founder & CEO | Quantitative finance, stochastic modelling, open science |
| D+A Strategies | Quant Research Trainee | Quantitative research, portfolio analytics, financial modelling |
| Stage | Objective |
|---|---|
| Research | Study financial markets, mathematical models, empirical phenomena and scientific literature |
| Modelling | Develop stochastic, statistical, probabilistic and machine learning models |
| Analysis | Quantify uncertainty, detect structure, estimate risk and validate assumptions |
| Impact | Translate research into reproducible tools, insights and decision frameworks |
| Output | Type | Status | Persistent identifier |
|---|---|---|---|
| Bachelier’s Theory of Speculation Revisited: A Reproducible Reconstruction of the Origins of Quantitative Finance | Preprint | Published on Zenodo | DOI: 10.5281/zenodo.21385499 |
| Deep Hedging under Transaction Costs: An Auditable NumPy Implementation and Exploratory Study | Preprint | Published on Zenodo | DOI: 10.5281/zenodo.21519919 |
| asr-deep-hedging v0.2.0 | Open-source software | Latest GitHub release | Repository · PyPI |
| Alpha Stochastic Research: Open Research and Reproducibility Framework | Research report | Published on Zenodo | DOI: 10.5281/zenodo.21379982 |
| asr-open-sc v0.3.2 | Open-source software | Released and archived on Zenodo | DOI: 10.5281/zenodo.21382430 |
Publication status: The Bachelier and Deep Hedging preprints are publicly available on Zenodo. The Deep Hedging implementation is maintained as an open-source Python package with a documented public API. No SSRN identifier is claimed until a public SSRN record exists.
Deep Hedging under Transaction Costs: An Auditable NumPy Implementation and Exploratory Study
An open-source and auditable NumPy implementation for neural option hedging under discrete rebalancing and transaction costs.
python -m pip install --upgrade asr-deep-hedgingfrom deep_hedging import (
Adam,
TanhMLP,
black_scholes_delta,
evaluate_positions,
simulate_gbm,
train_step,
)|
Published reproducible reconstruction of Louis Bachelier’s foundational work on financial speculation, arithmetic Brownian motion and option pricing. The preprint is archived on Zenodo, while the accompanying Python package remains under active maintenance and development. Areas: 
|
A rigorous framework for measuring extreme tail risk and regulatory compliance through VaR and CVaR methodologies. Areas: 
|
|
Research on probabilistic dependence, uncertainty and risk structures in cognitive offloading and artificial intelligence systems. Areas: 
|
Study of behavioural segmentation, equilibrium structures and probabilistic dynamics in institutional environments. Areas: 
|
|
Research on the relationship between naive diversification, optimized allocation, risk concentration and portfolio construction. Areas:
|
Implementation and analysis of hierarchical portfolio construction methods based on clustering, covariance structure and risk allocation. Areas:
|
|
Developed an auditable NumPy framework for discrete-time neural option hedging under transaction costs. The project includes exact empirical CVaR optimization, GBM and Heston simulation, manual network gradients, inventory-aware policies, classical benchmarks, evaluation utilities, automated tests, and a documented public Python API. Areas:
|
Research on agentic trading agents, validation frameworks, market decision systems and financial regulation constraints. Areas: 
|
| Mathematics | Finance | Artificial Intelligence |
|---|---|---|
| Probability Theory | Quantitative Finance | Machine Learning |
| Stochastic Processes | Portfolio Optimization | Deep Learning |
| Bayesian Statistics | Risk Management | Financial AI |
| Optimization | Derivatives Pricing | Scientific Computing |
| Time Series Analysis | Financial Econometrics | Reproducible Research |
| Numerical Methods | Market Modelling | Model Validation |
Alpha Stochastic Research (ASR) is an independent quantitative finance research laboratory dedicated to rigorous, transparent and reproducible research.
ASR focuses on:
- Quantitative finance
- Mathematical finance
- Stochastic modelling
- Financial machine learning
- Quantitative risk
- Portfolio optimization
- Scientific computing
- Open research and education
| Project | Description | Status | DOI / Repository |
|---|---|---|---|
| ASR Open Research and Reproducibility Framework | Institutional framework for transparent, reproducible and open quantitative research | Published | 10.5281/zenodo.21379982 |
| asr-open-sc | Shared registry and infrastructure for the ASR open-science Python ecosystem | v0.3.2 released | 10.5281/zenodo.21382430 |
| ASR Theory of Speculation | Reproducible implementation and modern reconstruction of Bachelier’s 1900 theory | Preprint published · package active | 10.5281/zenodo.21385499 |
| ASR Organization Profile | Community health files and institutional profile for Alpha Stochastic Research | Active | GitHub organization |
| VaR versus CVaR Framework | Extreme tail-risk and regulatory risk measurement framework | Active research | DOI pending publication |
| Portfolio Optimization Engine | Systematic allocation and portfolio construction research | Planned | Not yet published |
| Hierarchical Risk Parity | Risk-based portfolio allocation and clustering methods | Research | Not yet published |
| ASR Deep Hedging | Auditable NumPy framework for neural option hedging under transaction costs, empirical CVaR optimization, GBM/Heston simulation, benchmarks, and evaluation | v0.2.0 released · preprint published | Paper · Repository · PyPI |
| Cognitive Offloading Risk | Probabilistic framework for AI and cognitive-risk modelling | Research | Not yet published |
| Agentic Trading Systems | Research framework for autonomous trading systems | Draft | Not yet published |
| Channel | Current status and purpose |
|---|---|
| Zenodo | Active public archive for ASR papers, reports and software releases. Visit the ASR community. |
| GitHub | Active source for reproducible code, notebooks, documentation, issue tracking and open-source research. |
| ORCID | Persistent researcher identity: 0009-0006-0745-0380. |
| SSRN | Planned dissemination channel for working papers and preprints. No SSRN identifier is listed until a record is public. |
| Alpha Stochastic Research | Official laboratory website, research announcements and project releases. |
| Substack | Research notes and explanatory articles. |
| Medium | Public-facing quantitative-finance articles. |
| Discord | Community collaboration and discussion: join the ASR community. |
|
Research Profile 
|
Current Focus 
|
| Principle | Commitment |
|---|---|
| Mathematics | Build on rigorous theoretical foundations |
| Reproducibility | Make methods, code and results transparent |
| Open Science | Share knowledge through public research outputs |
| Engineering | Transform theory into usable quantitative tools |
| Integrity | State assumptions, limitations and uncertainty clearly |
| Continuous Learning | Maintain curiosity across finance, mathematics and artificial intelligence |