Time Series Analysis and Forecasting

Title Time Series Analysis and Forecasting
Quarter Summer 2017 (July 2017)
Instructor Dr Yuri Balasanov (
Syllabus Course Description
This is a fast paced 12 week course in Time Series Analysis for individuals with background in Linear Algebra, Calculus, Statistics and Probability. The course will introduce concepts like autocorrelation, stationarity, cointegration and more. The course will use time series concepts for developing sophisticated predictive models for financial markets. This course will use R for all programming assignments.

Course Contents
  • Stationarity
  • ARMA, ARIMA and GARCH models
  • Introduction to Spectrum
  • GMM
  • Introduction to VARs
  • Factor Models
  • Empirical Processes
  • Unit Roots
  • Cointegration
  • Filtering, State Space Models
  • Kalman Filter
Course Objectives:
At the completion of the course, students will be able to do the following:
  • Understand Stationary and non Stationary processes
  • Check data for autocorrelation with different lagged parameters
  • Find cointegrated pairs or baskets
  • Apply Time Series concepts to build predictive models
Instruction Format Coursework will have following four important components:
  • Weekly tasks
    • The course instructor will provide reading material, short videos explaining key concepts and lecture notes to be completed at home
    • Instructor will hold regular video conferences to go over concepts where the students need help
  • Weekly Sessions with Teaching Assistants. The purpose of these sessions will be:
    • Discussion on topics from the week
    • Working in small groups on a problem and presenting solution to the class. Other groups will be required to challenge the concepts and methodology used for problem solving
    • Working on group and individual projects
    • Assignment discussions
    • Quizzes and Exams
  • Reading Assignment and homework for the week
  • Virtual office hour with TA/Instructor via video conferencing
Assessment A letter grade A,B,C,D or F for the course will be decided based on

Projects: 40% of the final grade

  • 2 group projects and 2 individual projects
Mid Term Exam: 10% of the final grade
  • 30 minutes duration which will include both multiple choice and subjective problems
Final Exam: 15% of the final grade
  • 30 minutes duration which will include both multiple choice and subjective problems
Homework: 20% of the final grade
  • There will be 6-8 homework which will be manually graded and feedback will be provided
Quizzes: 10% of the final grade
  • There will be 4 quizzes which will have multiple choice format
Class Participation:10% of the final grade
  • Your participation will be evaluated based on lab discussion, questions/comments, replies on the discussion forum and teamwork on the group projects
TextBook Time Series Analysis by James D Hamilton
Pre-Requisite Prior experience with R, strong background in Statistics and Probability
Time Lecture Time: 8:00 pm – 9:30 pm EST, Tuesday evenings
Lab Time: 8:00 pm – 9:30 pm EST, Sunday evenings
Virtual Office Hour time:TBA
Location   Online
TA Information TBA
Effort Required 6-10 hours per week
Certification Participants will receive an instructor-signed certificate with a Pass grade if they score 50% and will receive Pass with distinction above 80%