Presented By O’Reilly and Intel AI
Put AI to Work
April 29-30, 2018: Training
April 30-May 2, 2018: Tutorials & Conference
New York, NY

Automatic financial econometrics with AI

Ambika Sukla (Morgan Stanley)
2:35pm–3:15pm Tuesday, May 1, 2018
Implementing AI, Models and Methods
Location: Nassau East/West
Average rating: ****.
(4.00, 2 ratings)

Who is this presentation for?

  • Data scientists, machine learning engineers, quants, strats, and hedge fund managers

Prerequisite knowledge

  • A basic understanding of machine learning, gradient descent, and probabilistic modeling
  • A background in finance (useful but not required)

What you'll learn

  • Learn how to break down very hard problems that are typically solved with advanced mathematics and automate AI techniques

Description

Financial econometrics is the science of building financial models. These models are developed for optimal trading strategies, risk management, portfolio optimization, market making, and asset pricing, among other tasks. Several decades of research from disciplines such as statistics, economics, operations research, and game theory have been applied to get to where we are today. However, today’s models make many assumptions to allow appropriate mathematics to be applied to derive realistic models.

Recent developments in AI have a huge potential to transform financial model building by automatically creating the models from data. Ambika Sukla covers the challenges in financial data and demonstrates how to apply deep generative models, reinforcement learning, and gradient descent optimization on problems such as time series analysis, volatility models, and high-frequency limit order book analysis.

Photo of Ambika Sukla

Ambika Sukla

Morgan Stanley

Ambika Sukla heads Morgan Stanley’s AI and ML Center of Excellence, where he works on applying machine learning techniques to algorithmic trading, risk management, operations and compliance, and wealth and investment management and helps set the firm’s AI strategy. He has extensive experience in machine learning, including recommendation systems, classification and regression, clustering, anomaly detection, and optimal control. Ambika is a big proponent of unsupervised and semisupervised learning methods. His background is in signal processing and information theory. He holds a master’s degree in telecommunication engineering from NJIT.

Comments on this page are now closed.

Comments

Jonathan Bates | MANAGER - RISK MANAGEMENT
05/04/2018 6:39am EDT

Hi, @Ambika,
Will the slides be made available? Great talk! Thank you!