Active Portfolio Management with Machine Learning and Time Series Forecasting (GY - ONLY)

  • Exploring and discovering new methods to optimize portfolio allocation using artificial intelligence and machine learning techniques

Active Portfolio Management with one black and one white bull intertwined.

Course Description

This VIP project is intended to explore and discover new methods to optimize portfolio allocation using artificial intelligence and machine learning techniques such as hidden Markov models, Kalman filters, random forests, genetic programming, deep learning, reinforcement learning, etc. This project is an opportunity to apply the ‘learning by doing’ methodology that starts with research and continues with engineering, hence simulating a research and development (R&D) environment. Students will continue working on a project developed by previous VIP groups. One of the central components of a VIP is that the project work builds beyond the participation of any one student, and students participate in something that could someday be very substantial.

As this VIP focuses on the development and the application of machine learning models in finance, students are required to have taken at least one machine learning or database systems course or be able to justify the mastery of one of these subjects prior to enrolling in the VIP project.

This is a 1.5 credits course (or optionally a 0 credit course) for graduate students only (MS and PhDs). This VIP project simulates the work of a quant researcher or a R&D engineer in finance and tech companies. Students are expected to deliver working code, and actively participate in the team effort. 

Students will be required to:

  1. Pair with a teammate to give a presentation on a topic of their choice related to the VIP theme.
  2. Contribute to the team implementation of a research paper and improve upon it along with the associated data science, modeling, and backtesting. 
  3. Contribute to the team summary report.

This opens the possibility for interested students to participate in competitions (e.g. the International Association for Quantitative Finance (IAQF) Student Competition, the Rotman International Trading Competition (RITC), etc.), present their work in workshops and conferences, and publish a research paper.

VIP Portolio Management Project

This project is intended to discover new methods to optimize portfolio allocation and evaluate them. The research team will be building machine learning models that take in historical time series data of the instruments, aggregate them into the ML pipeline, construct the portfolio, and rebalance the allocation regularly. The goal of this project is to build a prototype model that can rebalance a portfolio allocation for investors in real-time.

Research, Design, or Technical Issues Involved or Addressed

  • Research and Design Issues

    • Historical data and technical indicators to use
    • Appropriate models to use and compare their result
    • Evaluating the model performance and the associated risk
    • Building the model prototype
  • Technical Issues 
    • Exploratory data analysis
    • Data fetching and management
    • Machine learning pipeline building
    • Cloud platform application building
    • UI design


  • Algorithmic Research and Development

Majors and Areas of Interest

  • Financial Engineering
  • Mathematical Finanace
  • Finance
  • Data Science
  • Software Engineering
  • Trading Experience
  • Portfolio Management
  • Cloud Application Development
  • UI Design

Methods and Technologies

  • Active Portfolio Management
  • Real-time Trading
  • Data Science
  • Statistical Learning
  • Optimal Control
  • Machine Learning
  • Deep Learning
  • Reinforcement Learning
  • UI Design


  • NYU Tandon School of Engineering

Primary Instructors