UGA Bulletin Logo

Introduction to Computational Investing


Course Description

An introduction to implementing computational-based trading strategies from information gathering to market ordering and trading, including probabilistic machine-learning approaches to situational analysis and to trading decisions. We consider approaches like linear regression, decision trees, K nearest neighbors, and reinforcement learning and apply them to real- world trading.

Additional Requirements for Graduate Students:
Graduate students are expected to contribute more than undergraduates in class discussions. They will present additional technical papers and be assigned additional conceptual thought questions on exams. They are also expected to provide more depth in their answers on exams.


Athena Title

Computational Investing


Undergraduate Pre or Corequisite

CSCI 2720 or CSCI 2725


Semester Course Offered

Not offered on a regular basis.


Grading System

A - F (Traditional)


Course Objectives

As part of this course, students: 1. Will learn about financial time series data and how to manipulate and visualize it. 2. Gain exposure to the principles of computational finance and financial markets. 3. Implement two different significant machine-learning algorithms from scratch. 4. Apply machine-learning algorithms to financial time series data in order to develop trading strategies. Educational Outcomes Upon successful completion of this course, students should be able to: 1. Understand modern stock market mechanics and how they affect trading strategies. 2. Understand the computational infrastructure of hedge funds and institutional trading firms. 3. Manipulate and visualize financial time series data. 4. Create programs that leverage time series data to generate trading strategies. 5. Create machine-learning algorithms implementing stock trading strategies.


Topical Outline

Reading, slicing, and plotting stock data Working with many stocks at once Statistical analysis of time series Incomplete data Histograms and scatter plots Sharpe ratio and other portfolio statistics Optimizers: How to optimize a portfolio How Machine-Learning is used at a Hedge Fund Regression From a regression tree to random forest Assessing a Machine-Learning algorithm Ensemble learners, bagging, and boosting Assess a portfolio Market mechanics What is a company worth? The Capital Assets Pricing Model (CAPM) Technical analysis Build and assess a Random Forest learner Generate data sets that confound regression learners Dealing with data The Efficient Markets Hypothesis The Fundamental Law Portfolio optimization and the efficient frontier Building a market simulator Reinforcement Learning Q-Learning Dyna-Q ML methodologies for time series data Building an ML-based forex strategy Black Scholes Options Model