Indian Institute of Management, Udaipur
Post-Graduate Programme in Management
Year 2020-21
Course Outline for “Business Forecasting”
Credits :4 Term : VI
Instructor : Professor V. Nagadevara
Email :
[email protected]------------------------------------------------------------------------------------------------------
Course Objectives:
Forecasts may be either subjective or objective. The objective approach to forecasting
involves developing a model which is based past relationships between the variable to
be forecast and the factors that are likely to affect it. Objective forecasting methods
have several advantages over the subjective forecasting. They are not affected by what
the subjective beliefs and nuances of the forecaster. Some of the objective forecasting
methods can even include processes where the forecasting model learns from its past
errors. In addition, these methods provide an objective basis for evaluating forecast
accuracy and for developing confidence intervals for forecasts. This course
concentrates on these objective methods of forecasting.
The course intends to integrate theory and application. Theory is only justified by its
power of application in this course. This course focuses on the type of techniques
where the objective is to forecast the quantum of something, e.g., sales, customers
served, birth rates, or stock prices.
Risk and uncertainty are central to forecasting and it is generally considered good
practice to factor in the degree of uncertainty associated with forecasts and quantify it.
Consequently, an introduction to methods for distributional forecasting will be
provided.
Forecasting requires analysis of large amounts of data, both for building as well as
testing the models, and the required formulae and equations are often complicated.
Hence, the course uses statistical packages such as R and Excel based ForecastX.
The students, after completing the course, will be able to demonstrate knowledge and
understanding of:
different fields of application of forecasting techniques
the capabilities as well as limitations of quantitative-based forecasting methods
the importance of incorporating uncertainty in forecasting
communicating the forecast results effectively
Use of software packages for forecasting
Textbook:
1. “Forecasting and Predictive Analytics with ForecastX”, by Barry Keating, J. Holton
Wilson and Shovan Chowdhury, Tata McGraw-Hill, 7th Edition (SIE)
Suggested alternative books:
1. “Practical Timeseries Forecasting – A Hands-on Guide”, Galit Shmueli, 3rd Ed
(Special Indian Edition), 2018
2. “Forecasting Methods and Applications”, Spyros Makridakis, S C Wheelwright
and Rob J Hyndman, John Wiley and sons, 3rd Edition
Other materials as may be referenced during the course.
Course Structure: Presentation will be case based, and learning will be hands on. Each
technique will be first explained with a small dataset followed by a real life application.
Class coverage: Given below is a tentative schedule of topics which will be introduced
with practical applications drawn from different industries.
Session Topics to be covered Application Reading
No.
Session 1 Introduction to Business Forecasting Chapter 1
and discussions on syllabus,
evaluation components, general
guidelines, Overview of Forecasting
Perspective
Sessions Forecasting Long term Trends Using Transportation Chapter 1
2 and 3 Delphi, Qualitative techniques for Industry and
Forecasting Environment
Session 4 Measuring Forecast Accuracy,
Metrics for evaluating model
performance, Prediction Cones,
Prediction Intervals
Session 5 Forecasting Passenger Ridership - AmTrak Data Chapter 3
Decomposition of Time Series data;
Moving Averages and Exponential
smoothing
Session 6 Forecasting Consumer Sentiment Consumer Chapter 3
Index - Advanced methods in Sentiment
exponential smoothing Index
Session 7 Forecasting Mega Economic Trends - Chapter 6
Measuring Cyclical Component
Session 8 Forecasting Jewelry Sales with Jewelry Chapter 4
Multiple Regression methods
Sessions Forecasting with Cross sectional data Chapter 4
9 and 10
Sessions Forecasting Electrical Equipment Electrical Chapter 7
11 and 12 orders - Auto Regressive Models Equipment
Industry
Session Forecasting Australian Wines Beverage Supplemental
13 Forecasting with Artificial Neural Industry reading
Networks material
provided
Session Combining Forecasts: Case of Real Estate chapter 8
14 Houses sold
Session Integrative Case - Demand Food Industry Chapter 8
15 and 16 Forecasting for Perishable Short
Shelf-life Homemade Food
Session Forecasting Brand buying behavior - Consumer Supplemental
17 Marcovian Forecasting Models: Behavior and reading
Market share material
Using Markov Chains to Forecast
provided
Sales Booking
Session New Product Forecasting - Case: New Product
18 Metabilical: Pricing, Packaging and Introduction
Demand Foreecasting
Session Forecast Implementation Chapter 9
19
Session Project Presentations
20
Evaluation: The following will be the evaluation scheme
Component Weightage
Midterm 30%
Project (Group) 40%
Final Exam 30%
The Project will be a group submission. Apart from submitting a report, each group
will make a presentation on their project. It is encouraged that each group finds a rich
dataset for forecasting and work on it as the project. If a group cannot find an
interesting project to work on, it will be provided by the instructor. It will have a cost
associated with it.
The Final Exam will be during the end-term examinations week. There will be no
makeup for the Final Exam.
Class participation and attendance: You will find it extremely beneficial to complete
the reading assigned at the very least by the end of the day of class. It is assumed that
you will attend all classes. You should inform the instructor beforehand about missed
classes. The rules for missed classes are detailed in the student handbook provided by
the Programmes office.
Class preparation and conduct: It is presumed that you have access to a laptop
computer for carrying out the work. Generally, using laptop in class is not allowed,
unless instructed otherwise by the instructor, for some specific sessions. Use of mobile
phones in class is strictly not allowed. You should adhere to the code of conduct
documented by the students’ body. Plagiarism will be dealt with the strictest of
measures.
