Guillaume Weisang

Top Performing Data Scientist / Machine Learning Researcher / Ph.D.

image

Dynamic, highly motivated data scientist leader; Core experience managing and developing research teams focused on financial modeling and trading strategies. Published researcher on hedge fund models, factor selection, risk parity and Bayesian methodologies in academic journals and books. Extensive mathematical, statistical and computer science background.


Work Experience

Founder and Managing Director

Potenstats, Inc | Jan. 2019 - present

Delivers consulting services in finance-oriented data science.

  • Developed trading algorithms using Machine Learning and Reinforcement Learning (Tensorflow, PyTorch).
  • Delivered financial win models with ROI 85%+ (5-10% yield) for US and UK horse racing and NBA.
  • Developed and managed data scientists in career goal setting, statistical modeling concepts and applications, and day-to-day execution of workloads.
  • Developed and delivered R, python and C++ libraries.

Head of Data Science

Gambalyst / Alpha Peak | Sept. 2021 - Jan. 2023

Led a team of 6 data scientists in developing models and trading strategies for sport events at an international hedge fund.

  • Designed, developed and delivered probability models, infrastructure and trading strategies for Hong Kong, US and UK horse racing, NBA, and European soccer leagues.
  • Created over 10 financial win models for US and UK horse racing and NBA with aspirational yield of 5 to 10% for ROI above 85%
  • Coached and managed all 6 data scientists in career goal setting, statistical modeling concepts and applications, and day-to-day execution of their work.
  • Presented team results and findings to executives and company investors on a monthly basis.
  • Architected and setup the infrastructure, production environments, databases, python libraries for data ETL, and Machine Learning applications.
  • Designed and developed over 7 Python packages.
  • Implemented a development and production ecosystem using: OOP, Machine Learning, Docker containers, AWS, MongoDB, SQL, InfluxDB, Apache Airflow, MLflow, Spark (PySpark) and Optuna to support management of the financial models.

Senior Researcher - Data Scientist

SBIA Inc. | 2019 - Sept. 2021

Spear-headed a team of 4 data scientists developing sports betting models offered to high-end clientèle through a tout service.

  • Designed, developed and delivered over 12 models for NBA (Spread, Totals, Moneyline), MLB and NCAAMB capable of generating over 100 units of profits per season.
  • Lead researcher and final approver on vision and direction for all modeling.
  • Coached 4 data scientists in statistical modeling concepts.
  • Architected and setup the infrastructure, production environments, databases, python libraries for data ETL, and Machine Learning applications.
  • Designed and developed over 7 Python packages and Javascript.
  • Implemented a development and production ecosystem using: OOP, Machine Learning, AWS (S3, EC2, Sagemaker), MongoDB, SQL, Apache Airflow, MLflow, and Optuna to support management of the betting models.

Senior Researcher - Data Scientist

Crown Investment Group | 2017 - Sept. 2021

Led a team of 3 data scientists developing proprietary horse racing betting model.

  • Developed and implemented a proprietary wagering algorithm incorporating inputs uncertainty in R, C++ and Python.
  • Built data operational backend in MongoDB and Javascript.
  • Built the predictive framework ecosystem for over 80 models
  • Incorporated time series models using recurrent networks, convolutional neural networks and conditional logistic regression models built on Tensorflow, Keras, Scikit-Learn and AWS Sagemaker.

Advisor

True Bearing Insights | 2018

Advised hedge fund on developing Machine Learning/Data Science strategies to build their predictive models on shipping cargo contracts.

Assistant Professor of Finance

Clark University | 2011 - 2018

Full-time faculty member of the Graduate School of Management with an active research program and teaching master-level courses in statistics, computational finance and business analytics.

  • Conducted 12 extensive research projects in quantitative asset management and statistics of which 4 have been published in professional finance or statistics publications.
  • Taught 4 to 6 graduate advanced statistics and computational finance courses per year to over a 100 students per semester (including Matlab programming).
  • Published 4 manuscripts and produced 3 computing libraries in Gauss and Python about model selection and hedge fund replication.
  • Actively participated in committees for faculty compensation, hiring, research and program curriculum management.

Researcher - Consultant

Cargometrics Technologies, LLC | 2015

Developed machine learning algorithms to help build trading signals from proprietary data.

Instructor

Bentley University | 2009 - 2011

Taught courses in the departments of Mathematical Sciences and Finance.

  • Taught Financial Markets and Investments, Business Statistics, Applied Business Statistics to 30-40 students per semester.
  • Published two manuscripts on time series and quantitative due diligence using the Madoff scandal as a case study.

Information

Refereed Articles

  • 2019 · A Multi-country Study of Factors Influencing Expatriate Career Intentions, A. Joardar, G. Weisang, Journal of International Management, Volume 25, Issue 2
  • 2016 · Risk Parity Portfolios with Risk Factors, T. Roncalli, G. Weisang, Quantitative Finance, Volume 16, Issue 3, pp. 377-388.
  • 2011 · Tracking Problems, Hedge Fund Replication and Alternative Beta, T. Roncalli, G. Weisang, Journal of Financial Transformation, Volume 31, pp.19-29.
  • 2008 · Vagaries of the Euro: An Introduction to ARIMA Modeling, G. Weisang, Y. Awazu, Case Studies in Business, Industry and Government Statistics, Volume 2, Issue 1, pp.45–55.

Refereed Chapters

  • 2014 · Factor Selection In Dynamic Hedge Fund Replication Models: A Bayesian Approach, G. Weisang, Advances in Econometrics, Volume 34, special issue on Bayesian Model Comparison.
  • 2009 · Risk Management Lessons from Madoff fraud, P. Clauss, T. Roncalli, G. Weisang, International Finance Review, Volume 10, Chapter 17, Eds. J.J. Choi and M. Papaioannou, pp. 505-543.

Edited Chapters

  • 2017 · Risk Measurement and Management for Hedge Funds, G. Weisang. Hedge Funds: Structure, Strategies and Performance, Chapter 16, Eds. H. Kent Baker and Greg Filbeck, Oxford University Press.

Working Papers

  • 2015 · Asset Management and Systemic Risk, T. Roncalli, G. Weisang, available at ssrn.com
  • 2009 · Exploring non linearities in Hedge Funds: an Application of Particle Filters to Hedge Fund Replication, T. Roncalli, G. Weisang.
  • 2009 · A GAUSS Library for Particle Filters, T. Roncalli, G. Weisang.
  • 2008 · ACD models: Models for data irregularly spaced in time. A Gentle Introduction, G. Weisang

Projects

PF

Gauss · Open Source

a Gauss library written with T. Roncalli for particle filters (Generic Particle Filter; Regularized Particle Filter; Sampling Importance Resampling; Sampling Importance Sampling; Particle smoother).

aacd

R · Open Source

routines for the estimation and simulation of (Augmented) Autoregressive Conditional Regression models.

pyMinimax

Python · Open Source

a library for minimax filters (Work in progress)