Management 67000 – Business Analytics
Syllabus: Fall 2011

________________________________________________________________________________________________________________________________________________________________________________________________________
Section(s)  Instructor              O ffi ce        O ffi ce hou rs          C ontact Info.

1          Prof. Arnab B isi        K RA N 506    Tu esday 1-2pm         abisi@ purdue.edu
                                                W ednesday 11am -12pm Ph : (76 5) 494-4416
                                                Friday 2-3pm

2          Prof. Hui Zhao          K RA N 522    M onday 1:30-2:30pm    zhaoh@ purdue.e
                                                T hursday 1:30-2:30pm   Ph : (76 5) 494-3447
                                                or by appointm ent

3 and 4    Prof. Mohit Taw arm alani R AW LS 4019  Tu esday 11am-12pm     m t&
                                                W ednesday 11am -12pm Ph : (76 5) 496-2620
                                                Friday 2:30-3:30pm
                                                or by appointm ent

____________________________________________________________________________________________________________________________________________________________________________

Course Website: http://katalyst.mgmt.purdue.edu/.
____________________________________________________________________________________________________________________________________________________________________________

Section             In structor                        L ecture Time               Venu e

1                  Prof. A rnab B isi                 9:50-11:20am                RA W L 2082
2                  Prof. H ui Zhao                   11:30am -1:00pm             RA W L 2082

3                  Prof. M ohit Taw arm alani          1:10-2:40pm                 RA W L 2082

4                  Prof. M ohit Taw arm alani          2:50-4:20pm                 RA W L 2082

____________________________________________________________________________________________________________________________________________________________________________

Introduction

Data analysis and modeling are important skills for effective managerial decision making in business and industry. Advances in technology (computers, scanners, cell phones) have made significant amount of data available to managers. For example, the Dow Jones Industrial Average is one of the best-known and most widely watched indicators of the direction in which stock market values are heading. Administration and Congressional policymakers rely on statistics for budget decisions and related fiscal policy choices. The Federal Reserve System bases the monetary policy on data analysis. A manager needs to know if the manufacturing process is producing a quality product based on monitoring and assessing process performance. A sales manager has to develop tools to regularly monitor the performance of sales force. A manufacturer of certain electronic products needs to produce a forecast of future sales in order to decide whether or not to expand production. Banks use customer data to identify and design lucrative banking products. These are a few of the many examples from business where statistics can improve company performance.

The techniques learnt in this course will help you infer data and as such make better informed decisions. The course covers basic probability, decision analysis, statistical analysis (hypothesis testing and regression analysis), simulation and provides an introduction to optimization techniques. Probability models provide tools to handle uncertainty and risk. Statistical analysis focuses on the presentation of data and techniques to draw useful and valid inferences from data. Optimization models and decision analysis focus on techniques that use data to inform decision-making.

Course objectives

The course emphasizes applications of data analysis through cases and computer exercises. The focus of the course is as much on modeling and presenting solutions to business problems as on understanding statistical methods. Areas covered by the course include descriptive statistics, exploratory data analysis, probability, simulation, decision analysis, estimation, hypothesis testing, regression analysis, and optimization. At the end of the course, you should have

Methods of Instruction

The course is taught using a mix of lectures, case/data presentations, class discussions, interactive problem-solving, and computer exercises/demonstrations. Case/data set analyses provide opportunities for the creative application of statistics to unstructured management problems and also serve to illustrate the applications of statistics in managerial decision making. Homework exercises emphasize the understanding of concepts and aim to develop proficiency in translating business problems into statistical questions. Computer exercises familiarize the students with statistical software for data analysis.

Textbook and Course Materials for Management 670

Statistics for Business and Economics, 11th edition (revised) by D. R. Anderson, D. J. Sweeney, and T. A. Williams, South-Western Publishing, 2011. (SBE)
Course packets to be distributed

Software for Management 670

Minitab
Excel add-ons Solver and Data Analysis
Decision Tools (Excel add-ins @Risk, PrecisionTree, etc.)

Your role in the course

Preparation Each student is expected to be prepared for each class, to contribute to class discussions, and to complete all assigned readings and exercises. Lecture material will be posted on the course website. Students are expected to be aware of the announcements, events, and files posted on the course website. The class will be divided into several teams. All assignments will be handed in as a team. Class and team-work is subject to the following rules:

Proficiency in working with data and statistical modeling can be attained only through extensive practice with textbook problems and cases. It is recommended that students attempt the numerous exercises in the textbook on their own. Homework and case assignments must be turned in by 9:50 am on the specified due dates using Katalyst. The name of the submitted file should be HW_XXX_Team_YYY (or Case_XXX_Team_YYY), where XXX is the homework number and YYY is the team number.

Methodology of Case Analysis: Identify the analysis questions. Convert them to statistical inquiries. Perform a descriptive analysis of the data. Identify unusual data points. Formally analyze the problem using statistical procedures. Pertinent questions, and not the ability to conduct sophisticated statistical analysis procedures, should drive the choice of procedures. State the assumptions of the chosen statistical procedure and, if possible, verify them using the data set. Derive reasonable conclusions and recommendations that are supported by formal analysis. Identify unexplored questions and create a case report documenting your findings.

Format of Case Report: The write-up should not exceed six one-sided letter-size pages, including visual material such as charts, graphs, and tables. The reports must be typed using font sizes of at least 11 points, and margins of at least one inch on each side. Readability is of paramount importance. Adherence to type size and margin requirements is important to ensure fairness across teams. A typical report contains the following sections: (1) introduction and problem statement, (2) summary of results, (3) analysis: technical and non-technical, (4) recommendations, and action plans/suggestions for future study, and (5) all attachments (graphs, tables, etc.). Even though some of the analysis is technical and is presented as such, the results must be summarized and interpreted so that they are accessible to a non-technical audience. More specific expectations of the casereport may be provided with each case. In case of conflict with the above, the guidelines included with your case should be followed.

Midterm: The midterm exam is optional. If you do not take the midterm, the final exam will count for 50% of the grade. Anyone who takes the midterm may specify on the final exam if they would like the midterm grade to be counted or if they would like to transfer the weight of the midterm to the final exam.

Course Honor Policy

We expect and encourage students to discuss readings, case materials, and the concepts covered by the course with one another. However, do not falsely represent someone else’s work as if it were your own. Students are expected to prepare case, homework, or other assignments without the assistance or reference to students who have taken the class before, prior semester’s class notes and the like. Further, the use of Internet for finding solutions to cases and problems is prohibited. Specifically, students are expected to follow Purdue regulations governing student conduct (see http://www.purdue.edu/univregs/pages/stu_conduct/stu_regulations.html).While working individually and with your team members:

Each team member is expected to understand, for each homework and case assignment, the reason for choosing a particular technique, the mechanics of the solution procedure, and the implication(s) of the final solution(s) and recommendation(s).

Grading Policy/Rules

The following table lists the percentage of points allocated to each graded activity.

Homework assignments 10%
Case assignments 20%
Class and Team participation 20%
Midterm 20%
Final examination 30%

Dealing with Campus Emergency

In the event of a major campus emergency, course requirements, deadlines, and grading percentages are subject to changes that may be necessitated by a revised semester calendar or other circumstances. Katalyst and my email address are ways to get information about changes in this course.

Tentative Course Outline

__________________________________________________________________

Session

Class Topics (Topics with * are not covered in SBE. Use powerpoint notes.)

Pre-Session

Data Elements, Variables, and Observations

  • Qualitative and Quantitative Data
  • Cross Sectional and Time Series Data
  • Data Sources

Descriptive Statistics: Tabular and Graphical Methods

  • Frequency Distributions
  • Bar-Graphs, Pie-Charts, and Histogram
  • Cumulative Distributions and Ogive
  • Crosstabulations and Scatter Plots

Descriptive Statistics: Numerical Measures

  • Measures of Central Tendency
  • Measures of Dispersion
  • Box and Whisker Plot
  • Coefficient of Variation
  • Covariance, Correlation
  • Simpsons Paradox
  • Chebyshevs Theorem

Reading: SBE Chapters 1, 2 and 3; M670_0 (Data).ppt

1 ( 8/22)

Course Introduction

  • Data Sources
  • Descriptive versus Inferential Statistics

Descriptive Statistics

  • Box and Whisker Plots
  • Coding Categorical Data
  • Outlier Detection
  • Scatter diagrams
  • Covariance, Correlation

Reading: SBE Sections 3.1 – 3.5 (Appendix 3.1, 3.2); M670_1 (Descriptive Statistics).ppt

2 (8/23)

Probability

  • Contingency Table and Revision of Probability Estimates
  • Experiments, Outcomes, and Events
  • Addition Law
  • Joint and Marginal Probabilities
  • Conditional Probability

Reading: SBE Sections 4.1 4.4; M670_2 (Probability).ppt

3 (8/25)

Probability

  • Independence of Events
  • Updating Probability (Bayes Theorem)

Reading: SBE Sections 4.4, 4.5; M670_2 (Probability).ppt

Decision Analysis

  • Decision Problem
  • Decision Criteria
  • Decision Tree

Reading: SBE Sections 21.1 21.3; M670_3 (Decision Analysis).ppt

4 (8/29)

Decision Analysis

  • Value of Information
  • Using PrecisionTree1

Reading: SBE Sections 21.4; M670_3 (Decision Analysis).ppt

Discrete Probability Distributions

  • Random Variables
  • Expected Value, Variance, and Covariance
  • Binomial Probability Distribution

Reading: SBE Sections 5.1 5.4; M670_4 (Discrete Distribution).ppt

5 (8/30)

Discrete Probability Distributions

  • Poisson Probability Distribution
  • Linear Combinations of Random Variables*
  • Return/Risk Analysis of a Portfolio of Investments*

Reading: SBE Section 5.5; M670_4 (Discrete Distribution).ppt

6 (9/1)

Continuous Probability Distributions

  • Normal Probability Distribution
  • Standardized distributions
  • Examples: Portfolio returns, Stock-out/ inventory management*

Reading: SBE Sections 6.1, 6.2 (Appendix 6.1, 6.2); M670_5 (Continuous Distribution).ppt

Due (9:50 am): H/W 1 to be announced

7 (9/6)

Case Discussion: Forward Software, Inc.

Continuous Probability Distributions

  • Assessing Normality
  • Normal Approximation of Binomial

Reading: SBE Section 6.3; M670_5 (Continuous Distribution).ppt

8 (9/8)

Sampling Distributions and Estimation

  • Sampling Methods
  • Sampling Distributions and Central Limit Theorem
  • Sampling Distribution of a Sample Mean
  • Sampling Distribution of a Sample Proportion
  • Point Estimators and their Properties

Reading: SBE Sections 7.1 – 7.8 (Appendix 7.1, 7.2); M670_6 (Estimation).ppt

9 (9/12)

Interval Estimation

  • t-distribution
  • Confidence Interval for a Sample Mean
  • Confidence Interval for a Sample Proportion

Reading: SBE Sections 8.1 – 8.4; Appendix 8.1, M670_6 (Estimation).ppt

10 (9/13)

Simulation*

  • Sampling a Discrete Random Variable
  • Sampling a Continuous Random Variable: Inverse Transform Sampling
  • Monte Carlo Simulation and Latin Hypercube Sampling
  • Simulation Modeling with EXCEL

Reading: M670_7 (Simulation).ppt

Due (9:50 am): H/W 2 – to be announced

11 (9/15)

Class canceled in lieu of exam

Exam

Midterm Exam – (2 hours duration, date and time to be announced)

12 (9/19)

Simulation

  • Simulation Modeling with @Risk
  • Simulating Correlated Random Variables

Reading: M670_7 (Simulation).ppt

Hypothesis Testing (I) about population mean

  • Decision Risks
  • Hypothesis Testing Steps
  • Single Process/Population Hypothesis Tests
  • z – Test, t – Test
  • p – value
  • One Tail versus Two Tail tests

Reading: SBE Sections 9.1 – 9.4; M670_8_1 (Hypothesis Testing I).ppt

13 (9/20)

Hypothesis Testing (I and II)

  • Tests about Population Proportion
  • Paired Sample test
  • Two Sample tests
  • Sampling Distribution of Sample Variance
  • F-Test for Comparing Variances

Reading: SBE Sections 9.5, 9.6, 10.1 – 10.3, 11.1, 11.2; M670_8_1 (Hypothesis Testing I).ppt; M670_8_2 (Hypothesis Testing II).ppt

14 (9/22)

Case Discussion: MEM: New Technologies

Simple Regression Models

  • Regression Modeling
  • Statistical Model and Assumptions
  • Least Squares Estimation
  • Coefficient of Determination

Reading: SBE Sections 14.1 – 14.4; M670_9 (Simple Regression).ppt

15 (9/26)

Simple Regression

  • Testing for Significance
  • Residual Analysis
  • Prediction

Reading: SBE Chapter 14.5 – 14.9; M670_9 (Simple Regression).ppt

Due (9:50 am): H/W 3 – to be announced

16 (9/27)

Multiple Regression Models

  • Interpretation of coefficients
  • Partitioning the Sum of Squares
  • F-test and t-test

Reading: SBE Sections 15.1 – 15.5; M670_10 (Multiple Regression).ppt

17 (9/29)

Multiple Regression Models

  • Partial F-test
  • Multicollinearity
  • Prediction

Reading: SBE Sections 16.2, 15.5, 15.6; M670_10 (Multiple Regression).ppt

18 (10/3)

Multiple Regression Models

  • Dummy Variables
  • Interaction Terms
  • Residual Analysis

Reading: SBE Sections 15.7, 15.8, 16.1, 16.4; M670_10 (Multiple Regression).ppt

19 (10/4)

Linear Programming*

  • Model Formulation
  • Graphical Solution
  • Types of linear programs
  • Spreadsheet Modeling

Due (9:50 am): H/W 4 – to be announced

20 (10/6)

Case Discussion: Newfood Corporation

Linear Programming Applications*

  • Transportation Problem
  • Financial/Inventory Planning (time permitting)
Exam

Final Exam: (4 hour duration, date and time to be announced)

Summary of Course Schedule

------------------------------------------------------------------------------------
 Class  D ate  T opic                  R eading                                   Due
------------------------------------------------------------------------------------
     1  8/22  Introd uction            S BE 3.1–3.5, A 3.1, 3.22;
--------------D-escriptive S-tatistics----M-670x1(D-escriptive ...).ppt---------------------
     2  8/23  P robability             S BE 4.1-4.4;
                                     M 670x2(Probability).ppt
------------------------------------------------------------------------------------
     3  8/25  P robability             S BE 4.4, 4.5, 21.1–21.3;
--------------D-ecision A-nalysis------M-670x3(D-ecision-...).ppt------------------------
     4  8/29  D ecision A nalysis      S BE 21.4, 5.1–5.4;
--------------D-iscrete-Distributions----M-670x4(D-iscrete-...).ppt-------------------------
     5  8/30  D iscrete Distributions    S BE 5.5;
                                     M 670x4(D iscrete ...).ppt
------------------------------------------------------------------------------------
     6   9 /1  C ontinuous             S BE 6.1, 6.2, A 6.1, 6.2;                H/W 1
--------------D-istributions------------M-670x4(C-ontinuous ).p-pt-----------------------
     7   9 /6  C ontinuous             S BE 6.3;                                Case 1
              D istributions            M 670x4(C ontinuous ).p pt
--------------S-ampling-Distributions--------------------------------------------------
     8   9 /8  S ampling Distributions   S BE 7.1-7.8, A 7.1, 7.2;

--------------P-oint E-stimation--------M-670x6(Estim-ation).ppt--------------------------
     9  9/12  Interval E stimation      S BE 8.1-8.4, A 8.1;
              S im ulatio n             M 670x6(Estim ation).ppt;
-------------------------------------M-670x7(Simu-lation).p-pt-------------------------
    10  9/13  S im ulatio n             M 670x7(Simu lation).p pt                  H/W 2
------------------------------------------------------------------------------------
----11--9/15--C-lass canceled-----------------------------------------------M-id-term--
    12  9/19  H ypothesis Testing I    S BE 9.1-9.4;
-------------------------------------M-670x8x1-(H-ypothesis ).ppt----------------------
    13  9/20  H ypothesis Testing I    S BE 9.5, 9.6, 10.1-10.3, 11.1, 11.&#
              and II                 M 670x8x1 (H ypothesis ).ppt

-------------------------------------M-670x8x2-(H-ypothesis ).ppt----------------------
    14  9/22  S im ple Regression       S BE 14.1-14.4;                          Case 2
-------------------------------------M-670x9-(Sim-ple Regression).ppt-----------------
    15  9/26  S im ple Regression       S BE 14.5-14.9;                          H/W 3
-------------------------------------M-670x9-(Sim-ple Regression).ppt-----------------
    16  9/27  M ultiple R egression     S BE 15.1-15.5;
                                     M 670x10 (M ultiple R egression).ppt
------------------------------------------------------------------------------------
    17  9/29  M ultiple R egression     S BE 16.2, 15.5, 15.6;
-------------------------------------M-670x10-(M-ultiple R-egression).ppt---------------
    18  10 /3  M ultiple R egression     S BE 15.7, 15.8, 16.1, 16.4;
-------------------------------------M-670x10-(M-ultiple R-egression).ppt---------------
    19  10 /4  L inear P rogram ming    M 670x11 (Linear Program ming).ppt         &
------------------------------------------------------------------------------------
----20--10-/6--L-inear P-rogram-ming----M-670x11(Linear Programm-ing).ppt--
    E      –                                                             Final E xam
------------------------------------------------------------------------------------

2A3.1 denotes Appendix 3.1

Installation of DecisionTools Student Edition

MGMT 670 uses several EXCEL Addins such as @Risk and PrecisionTree. These are available as part of DecisionTools suite from Palisade Corporation (http://www.palisade.com/decisiontools_suite/). As a student of Krannert, you can download the academic student version of DecisionTools suite on your personal computer/laptop. The software will expire in one year. In addition, this software is also available in the MBA Lab in RAWLS Hall. You may download the DecisionTools Suite from: https://www.krannert.purdue.edu/departments/kcc/resources/palisade/Palisade_v5.7_Student.zip.

If you are prompted to authenticate, and you should respond with:

U sernam e: krannert\< your career uid>

Password:  < your pw >

If you are on a computer that is already logged into Krannert domain, you will not be asked to authenticate again. If you have trouble downloading and installing the DecisionTools suite, please get in touch with KCC on 7th Floor in the Krannert Building (kcchelp@purdue.edu).

Installation of Minitab

MGMT 670 makes extensive use of MINITAB for regression analysis. The software is accessible in the MBA lab on the 4th floor of RAWLS Hall (Rawls 4082). You can lease or purchase a version for your personal computer at at http://www.onthehub.com/minitab/.