Devlina Chatterjee
Associate Professor
Department of Management Sciences
IIT Kanpur, 208016
devlina@iitk.ac.in

Home
     
Devlina Chatterjee - Home Page
     

MBA752
Time Series Modeling for Business Analytics
Course Syllabus
Semester: 2021 Fall
Timings: M, Th: 10:35 am – 11:50 pm,
Classroom: Online zoom meetings
Instructor: Dr. Devlina Chatterjee, Room 211, IME Building
Ph: 259 6960 (Office)
Email: devlina@iitk.ac.in
Objective of the course:

This is an introductory graduate level half semester course on forecasting and time series modeling. The focus of the course is to understand the applications of time-series model to actual business applications. The underlying statistical theory will be emphasized to the extent that it helps students set up an appropriate model.
 
 
Syllabus:
Understanding and decomposition of time-series data, unit root and stationarity, moving average and exponential smoothing models, times-series regression models (AR, MA, ARIMA), dynamic causal regression, vector auto-regression (VAR) and volatility modeling (ARCH / GARCH models).
Text Book:
Forecasting: principles and practice, RJ Hyndman, G Athanasopoulos 2018, https://otexts.com/fpp2
 
Reference Material:
  • The Analysis of Time-series: An Introduction with R by Chris Chatfield and Haipeng Xing
  • Analysis of financial time series / Ruey S. Tsay. 3rd ed. (Wiley series in probability and statistics)
  • Practical Time Series Forecasting with R: A Hands-On Guide [2nd Edition] (Practical Analytics) by Galit Shmueli, Axelrod Schnall Publishers
Evaluation scheme:
In order to benefit from this course, active participation is required from the students in classes and also out of classes doing assignments and project work.
Attendance* - 10%
Quizzes (2) - 30%
Term Projects (1) - 30%
Final exam - 30%
*(minimum of 65% attendance required - else deregistered from the course). Attendance mandatory on the days of project presentation
Academic Integrity:
If you are caught cheating or copying on any assignment, quiz or exam, you will
  1. get ZERO marks on that assignment/quiz/exam.
  2. an additional penalty may be given including and upto assigning a grade F in the course.
Week Lecture Topic
1 1 Getting Started, Introduction to Time-series data
1 2 Time-Series Graphics
2 3 Time-series decomposition
2 4 Forecaster’s toolbox
3 5 Forecaster’s toolbox
3 6 Time-series Regression Models
4 7 Exponential smoothing
4 8 Exponential smoothing
5 9 ARIMA Models
5 10 ARIMA Models
6 11 Dynamic causal regression models
6 12 Dynamic causal regression models
7 13 Advanced Forecasting methods
7 14 Advanced Forecasting methods