CS7646 Summer 2017 ATL
Contents
Overview
This summer, the course will follow this broad outline:
- Brief introduction to Manipulating Financial Data in Python
- Introduction to Machine Learning
- Computational Investing
- Machine Learning Algorithms for Trading
Instructor information
David Byrd
Research Scientist, Interactive Media Technology Center at Georgia Tech
2017 Summer Schedule
Class meets MW 12:30 - 2:20 in Klaus 1456
2017 Summer Schedule
This schedule is tentative and subject to change due to the compressed summer timeline. I am not certain exactly how quickly we will progress through the material.
The first exam will be given in week 6 or 7 depending on our progress. I will nail it down early in the class.
The second exam will be given on the final day of regular class in week 10.
| Week | Topic | Due |
|---|---|---|
| 1 | Course Overview, Intro to Markets, Intro to Machine Learning, Pandas Tutorial | |
| 2 | Visualizing Market Data, Pricing Stocks, Numpy Tutorial, Working with Time Series | |
| 3 | Incomplete Data, Plots, ML Lexicon/Taxonomy, Evaluating Learners | MC1-P1 |
| 4 | Supervised Learning (KNN, LinReg, Decision Trees, Boosting, Bagging) | |
| 5 | Market History, Actors, Order Book, Order Types | MC3-P1 + MC3-HW1 |
| 6 | Markets, Valuation, Capitalization, Time Value of Money | Exam 1 |
| 7 | Options, Leverage | MC2-P1 |
| 8 | Technical Analysis, Candlestick Chart Patterns, CAPM, Efficient Market Hypothesis | |
| 9 | Fund. Law of APM, Efficient Frontier, Finite Automata, Markov Decision Processes | MC3-P2 |
| 10 | Value Iteration, Q-Learning | Exam 2 |
| 11 | Final class, extra topics | Final project (MC3-P3) |
Assignments & Grading
This is a project-heavy class. There will be 5 projects this semester, due roughly every two weeks. Note that there are two non-cumulative exams during the semester. The final project in the class (and the most complex) will be due during the final exam period, in lieu of a final exam.
Assignments are available on the main course page and are the same for the on-campus and online class.
Participation: For the 3% participation credit, Piazza participation is not necessary for on campus students. Participation will be judged based on attendance and attention at the lectures.
Office hours
David Byrd (Instructor), MW 2:20-3:30, Klaus 1456. There is no class after ours in the classroom, so I will simply stay there for up to an hour every class, as long as anyone wants to talk to me!
TAs TBD. A poll was sent out for TA office hours via T-Square. Be sure to complete it!
