Course Description

In this course, the student will learn to identify, evaluate, and capture business analytic opportunities that create value for an organization. Theoretical data analytics methods, as well as case studies on successful analytics applications, will be covered. Basic descriptive analytics methods are reviewed along with a quick introduction to using Python in analyzing large data sets. Predictive analytics techniques including clustering, classification, and regression, are covered in detail. Prescriptive analytics applications on utilization simulation and optimization over large data to improve business decisions are presented. Case studies emphasize financial applications such as portfolio management and automated trading.

Pre-requisites

• By Course: STAT 230, INDE 301, INDE 302, INDE 303, INDE 504
• By Topic: Programming, Probability and statistics, engineering economy, optimization theory, stochastic processes, Monte Carlo simulations

Course Objectives

• A practical ability to carry out data analysis using computational tools on problems of applied nature.
• An understanding of the importance of data analytics in the decision making process of modern organizations.
• An appreciation of the challenges in applying data analytics in practice.
• An exposure to modern applications of data analytics, especially in Finance.
• An overview of the main predictive analytics tools such regression, classification, and clustering.

Learning Outcomes

• Analyze data sets with Python and perform basic descriptive analytics.
• Identify the suitable data analytics tools that assist organizations making data driven decisions.
• Understand and implement basic linear and nonlinear regression, clustering, classification and other predictive analytics techniques.
• Apply familiar prescriptive analytics tools such as simulation and optimization in large-data contexts.
• Utilize predictive and prescriptive analytics in modern applications, especially financial planning and trading.

Topics Covered

• What is Statistical Learning, assessing Model Accuracy
• Simple and Multiple Linear Regression
• Classification (Logistic Regression and Linear Discriminant Analysis)
• Re-sampling Methods (Cross-validation and Boostrapping)
• Linear Model Selection and Regularization
• Tree-Based Methods
• Support Vector Machines
• Unsupervised learning (K-means and Hierarchical Clustering, Principal Component Analysis)
• Case studies and applications

Software and Coding

• The class will mostly use Python 2.7, a powerful programming language commonly used in academia and business.
• This course is not a Python programing class. We will provide limited software instruction, in-class demonstration, and code to accompany lectures and assignments. Like any programming language, Python is best learned through practice.
• You should install Python as soon as possible and familiarize yourself with basic operations. We recommend to install Anaconda, a distribution containing Python and most of the packages required for data analysis. You can download it here. The more proficiency you can gain prior to class, the more you will get out of the sessions. There are many tutorials online such us this one.

Schedule

• Every Thursday, between 2pm and 5pm
• One lecture and one lab session each week.

Evaluation Method

• Exam: 40%
• Labs / Project: 40%
• In-class Quizzes: 15%
• Moodle Questionnaires: 5%

Contacts: teaching@algotraders.org