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

VIP graphic with a ladder as the right-half of the V in purple and green

This project is intended to discover new methods to optimize portfolio allocation. The research team will be building an application that takes in historical data of the instruments and forecast the return by machine learning and time series methods. Then, apply the aggregated pipeline onto the portfolio and rebalance the allocation regularly. The goal of this project is to establish an application that can perform real-time rebalance portfolio allocation for investors.

Research, Design, or Technical Issues Involved or Addressed

  • Research and Design

    • Portfolio instrument universe

    • Historical data and technical indicators to use

    • Appropriate models to use and compare their result

    • Building the application on a cloud platform for live rebalance (AWS)

    • Potential connection with already existing broker API (Alpaca)

    • Backtesting the result of the allocation

  • Technical issues involved

    • Explanatory data analysis

    • Data fetching

    • Machine learning pipeline building

    • Cloud platform application building

    • UI design

 

Subteams

There are three subteams under this project.

  • Data science team

    • In charge of the EDA process, forecasting the return, building models pipeline, comparing the result of each model

  • Trading team

    • Incorporate practical trading details into backtesting

  • Engineering team

    • In charge of building the pipeline onto the application platform, debugging the system

    • Design the UI that can present the application

    • Ensure connection with data and broker API

 

Reference

Merger & Acquisition Outcome Prediction

  1. CapitalVX: A Machine Learning Model for Startup Selection and Exit Prediction

  2. Nicholas Center’s First Ever Machine Learning Consulting Project – Research Report on Predicting M&A Targets

Active Portfolio Management with Machine Learning and Time Series Forecasting

  1. Intelligent Portfolio Construction: Machine-Learning enabled Mean-Variance Optimization

 

 

Subteams

  • Data Science
  • Trading
  • Engineering

Majors and Areas of Interest

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

Methods and Technologies

  • Active Portfolio Management
  • Real-time Trading
  • Machine Learning
  • Deep Learning
  • Amazon Web Services
  • Broker API
  • UI Design

Partners

  • NYU Tandon School of Engineering

Primary Instructors