QMBU 301 Quantitative Methods in Business
Spring 2009
Assistant Prof. Dr. Özden Gür Ali
CASE-241, Tel: 212 338 1450,
Email: , homepage:
Overview
Managing a business requires making decisions under uncertainty – be it annual budgeting/hiring, investment, or weekly production decisions. This course introduces students to quantitative methods that facilitate objective and rational managerial decision making under uncertainty: Regression and forecasting techniques help analyze the data about past/current similar situations. The identified patternsare projected onto new situations to enable business decision making and planning.
Covered methods are time series forecasting techniques, regression and an overview of qualitative methods.We focus on forecasting techniques that have been most useful in business and administration problems. These include moving averages, exponential smoothing, decomposition methods, transformations, and regression with time series (autoregression, multiple regression, nonlinear transformations). Other methods are mentioned along with their advantages and disadvantages.
These methods are applied in business situations from different industries and functional areas.Drawing conclusions based on these analyses is emphasized. Computer based tools are used extensively.
Objectives
This course is designed to develop the quantitative skills required to make effective business decisions under uncertainty, specificallyto
- identify the right method(s) for the situation
- apply the methods correctly
- use software tools to facilitate analyses
- identify and choose variables
- check for model appropriateness
- assess validity of analyses
Textbooks:
Main Textbook:
“Business Forecasting”, by J. Hanke and D. Wichern, eighth edition, 2005
Supporting Text:
“Forecasting – Methods and Applications”, by S. Makridakis, S. Wheelwright, R. Hyndman, third edition, 1998
Lectures:
Lectures will include the discussion of methods, examples with business data, short cases from different industries and functional areas. Students are expected to have read and thought about how to approach the short cases indicated on the syllabus before coming to class. Class participation in lectures and case discussions contributes to students’ grades. A guest lecturer from the industry will provide an overview of their forecasting efforts. Students are required to attend this lecture which will be scheduled at a different time, combining both sections.
Study problems and Quizes:
Quizzes will be given regularly. They will be based on voluntary homeworks, which you are not required to return for grading. Homework assignments focus on applying the methods, evaluating their adequacy and recommendation of a course of action based on analyses. Students are required to take the quizzes in the section that they are registered in (otherwise the quiz will not be graded). There is absolutely no make-up for the quizzes.
Exams:
There will be two midterms and one final exam. You should expect the make-up exams to be no easier than the regular exams.
Labs and Software:
Students will use Excel to apply the data analysis methods learnt in class. They will first work through a familiar problem under guidance and then work on a similar assignment independently. The evaluation of lab performance will be based on attendance, and performance on the assigned problem. The labs will be held in SOS 180.
Teaching Assistants: TBA
Grading:
Midterm 120%
Midterm 220%
Final exam 30%
Lab sessions15%
Quizzes10%
Participation 5%
Total100%
Office hours:TBA.
Course webpage: accessible from KUAIS
Remarks:
1. Academic dishonesty is considered unacceptable within the University community, and it includes and is not limited to the following in the KU Code of conduct:
Cheating includes, but is not limited to, copying from a classmate or providing answers or information, either written or oral to others, in an examination or in the preparation of material subject to academic evaluation.
Plagiarism is borrowing or using someone else’s writing or ideas without giving written acknowledgement to the author. This includes copying from a fellow student’s paper or from a text or internet site without properly citing the source.
Multiple submission includes resubmission of the same work previously used in another course or project, without the permission of the instructor for both courses.
Collusion is getting unauthorized help from another person such as having someone else write one’s assignment, or having someone else take an exam with false identification. Impersonating a student in an examination is also considered a grave act of dishonesty.
Fabrication includes, but is not limited to, falsification or invention of any information or citation in an academic exercise.Facilitating academic dishonesty includes, but is not limited to, knowingly helping another student commit an act of academic misconduct (e.g., cheating, fabrication, plagiarism, multiple submissions).
2. Students are responsible for all announcements made in class and on courseweb
3. The class-room rules of conduct apply. The activities which are prohibited in class include and are not limited to:
- Engaging in side conservations.
- Using cell phones and other electronic devices.
- Using laptops for purposes that are not course-related.
- Arriving late or leaving early without the prior permission of the instructor.
- Reading material, e.g., magazines, newspapers, novels etc., that are not course-related.
- Working on personal activities or the assignments of other courses.
- Interrupting the professor or other students.
- Trashing the classroom.
Tentative Schedule:
DateTopicReading
Feb.10Introduction
12Review of basic statistical conceptsCh2
Describing data
Random variables and probability distributions
17Sampling
Hypothesis testing
Correlation
Cross sectional versus time series data
19Simple linear regressionCh 6
24Evaluation of regression
26Transformations
27Lab exercise-1 Introduction to data analysis with Excel
Mar.3Case 6-6: AAA Washingtonp 260
Multiple regression analysisCh 7
5Inference
Model building
10Diagnostics
Case 7-1: The Bond Marketp 310
12Time series data patternsCh 3 (57-73),
Autocorrelation, differencing
Case 3-2: Mr Tuxp 92
13Lab exercise-2 Regression and transformations
17Naive models, Moving averageCh. 4 (101-113)
Evaluating the accuracy and adequacy of the forecast
19Midterm 1
20Lab exercise- 3 Time series patterns
24Exponential smoothing – SimpleCh. 4 (114-131)
Choosing a forecasting methodCh. 3 (74- 84)
26Forecasting methods with for data with trend, Holt’s
Case 4.3 Consumer Credit Union
28 Lab exercise- 4 Time Series Forecast Accuracy Measures
31Winter’s exponential smoothing
Apr.2Decomposition MethodsCh5
Case 5-4: Murphy Brothers
3Lab exercise- 5 Exponential Smoothing
14Trend, Deseasonalization
Random Walk Model
Case 5-2: Mr Tux
16Time Series forecasting exercises
18Lab exercise- 6 Decomposition
21Business/Economic Indicators(p176-177)
28Regression with Time Series Ch8
Durbin Watson test(p327-342)
30Case 8-7: Alomega Food Storesp 375
Autoregressive modelsCh8 (p345-346)
May1Lab exercise- 7 Regression with Time Series
5Case 8-3: Restaurant Salesp 367
Introduction to ARIMA ModelsCh 9 (p386-389)
7Midterm 2
12Judgmental Forecasting and AdjustmentsCh. 10 Combining forecasts
14Managing the forecasting processCh 11
Case 11-4: Mr Tuxp 500
21Review