FE 670 - Algorithmic Trading
 
Prof. Steve Yang
Stevens Institute of Technology
Fall Semester 2012

Instructor: Steve Yang, Babbio 536, steve.yang@stevens.edu

Class Time: Lectures on Tuesday 06:15PM-08:45PM (08-27-2012 – 12-21-2012)

Office Hours: Wednesday 10:00AM-11:00AM at Babbio 536

Lab Time:  Hanlon FSC LAB 8/28, 10/16, and 11/27

Prerequisites: FE545

Topics:  This course investigates methods implemented in multiple quantitative trading strategies with emphasis on automated trading and quantitative finance based approaches to enhance the trade-decision making mechanism. The course provides a comprehensive view of the algorithmic trading paradigm and some of the key quantitative finance foundations of these trading strategies. Topics explore markets, financial modeling and its pitfalls, factor model based strategies, portfolio optimization strategies, and order execution strategies. The data mining and machine learning based trading strategies are also introduced, and these strategies include, but not limited to, Bayesian method, weak classifier method, boosting and general meta-algorithmic emerging methods.

Textbook:  Frank J. Fabozzi, Sergio M. Focardi, and Petter N. Kolm, , Quantitative Equity Investing: Techniques and Strategies (Wiley, 2010), 2007, and Barry Johnson, Algorithmic Trading & DMA, 4Myeloma Press London, 2010..

Lecture Outline:

Week

Topic(s)

Reading(s)

HW

1

An Overview of Trading and Markets

Barry Johnson [1,2,3]

 

2

Common Pitfalls in Financial Modeling

Frank J. Fabozzi [4][1,2]

HW1

3

Factor Models and Their Estimation

Frank J. Fabozzi [5]

 

4

Factor-Based Trading Strategies

Frank J. Fabozzi [6-7]

 

5

Portfolio Optimization and Black-Litterman Model

Frank J. Fabozzi [8-9]

HW2

6

Robust Portfolio Optimization

Frank J. Fabozzi [10][3]

 

7

Transaction Costs & Trade execution

Frank J. Fabozzi [11]

HW3

8

Transaction Costs & Optimal Strategies

Barry Johnson[8,9]

HW3

Mid-term Exam

 

EXAM-I

10

Order Placement & Execution Tactics

Barry Johnson[10]

 

11

Enhancing Trading Strategies

Barry Johnson [10]

 

12

Pattern Recognition Models: Bayesian Networks, Hidden Markov Models

Academic papers[13]

HW4

13

Pattern Recognition Models: Decision Trees,

Academic papers

 

14

Final Exam

 

EXAM-II

 

Exams and Grades:

Assignment

Grade Percentage

Assignments

20%

Project

20%

Midterm exam

30%

Final Exam

30%

Total Grade

100%

 

 Exams: Two Exams. (Mid-term) EXAM I: Oct 23 - (Tues). (Final) EXAM II: Dec. 11 - (Tues). These exams will consist of short questions, and data analysis using R.

 

Exam must be taken at these times – No Exceptions!!!!!!!

 

Exam Honor Policy: You are not allowed to discuss any of the exam questions with one another or to show any of your solutions. The work must be done independently and pledged.

 

Homework: There will be 4 homework assignments (approx every 2-3 weeks).

 

Homework Honor Policy: You are allowed to discuss the problems between yourselves, but once you begin writing up your solution, you must do so independently, and cannot show one another any parts of your written solutions. The HW is to be pledged (that it adheres to this).

 

Final Project: You will be given a dataset, and you will apply the methods which you have learned. If you do it right, this can be an immensely satisfying experience. You will turn in this project - I don't want the computer output, but descriptions of the results IN YOUR OWN WORDS - 3 single spaced pages, including plots, at most. We will talk more about this as the semester proceeds. You will each give a brief presentation on your project to the class, during the last week - Attendance is MANDATORY at the presentations – Dec 18 (tentatively)!!!

 

Attendance: Attendance will be taken randomly (e.g., 6-7 times during the semester) and will determine "which direction" the resulting grade will “fall”, for those grades which are borderline (e.g., between B+ or A-).