Project

profileWorld123
GroupProject_Instructions1.docx

Group Project: Memo Business Analytics/Data & Decisions

This is a group exercise. Please put yourselves in groups of 4-6 students. Any groups outside of this size-range will receive a 10-point deduction . (This should not be an issue.)

To turn in the assignment, one group member should upload the memo to ICON’s Dropbox facility. Please upload a single file in PDF format by the due date/time noted on the dropbox. In particular, do not upload any Excel files. If you need to merge PDFs before submitting, try searching “merge PDF” on the web for free websites to do so.

DISCUSSION

You and your teammates work for a statistical consulting firm. You have recently been hired by the Des Moines Register to provide the background analysis for an upcoming news story on the housing market in Des Moines. Your contact at the Register is Jennifer Baumgartner, and she has asked you to perform a regression analysis fitting the value of single-family homes in Des Moines.

DATA

Using Polk County public records, you have collected information on a random sample of 750 single-family homes sold in Des Moines during the two-year period March 8, 2013 to March 7, 2015. The data can be found in the file “Data for Group Project.xlsx.” Please pay special attention to the exact definition of variables in the data.

REQUIRED ANALYSIS

Using the principle of parsimony, develop a best possible multiple linear regression model to fit the sale price of single-family homes in Des Moines and its western suburbs. Then analyze the quality of the model by answering the six key questions.

PLEASE BE SURE TO STATE WHAT YOU CHOSE AS YOUR INITIAL BASE CASES FOR ANY CATEGORICAL VARIABLES.

ASSIGNMENT

Incorporate the above analysis into a memo addressed to Ms. Baumgartner. The memo should begin with a brief executive summary (which summarizes your model and analysis in a few direct sentences) and then continue with the main body (which presents your model and analysis in detail and points out any possible limitations). In particular, you can be fairly technical in the main body, especially when answering the key questions. However, make every effort to show and explain the final regression equation (“what you tell the boss”) in terms that are easy for anyone to understand. The memo should be limited to a maximum of 5 single-spaced pages, including all figures, tables, appendices, and any other supporting materials. Margins should be 1 inch on all sides, and the font size should be 11 point or larger.

NOTES AND HINTS

Please imitate regression techniques and analyses that have been shown in class. For example, your models should use dummy/indicator variables for categorical data, and in the technical session, you should consider the “key questions” for evaluating the fit of a regression.

In general, you do not need to include in the memo every bit of information or Excel output encountered during your analysis, but if some portion of your analysis supports your findings, then that portion should be included.

Do not attempt to remove outliers, gather new data, or run a nonlinear regression. However, if you see something very strange in the data, please let the instructor know just in case he has made some kind of mistake in preparing the data.

Please see the page below for grading guidelines.

IMPORTANT

If you have any questions about this memo, please ask Ben. He is always willing to clarify. Please do not make assumptions about what is or is not required to achieve a good grade.

Note that while this is a quantitative project that has a correct ‘final answer’, some of the grading component will be subjective based upon the quality of analysis/explanations/report as compared to other groups.

Grading components will include, but are not limited to, the following:

· Executive Summary:

· This should effectively summarize your model and analysis in a few direct sentences without being overly technical.

· Analysis and main body of memo:

· Should correctly apply regression techniques to arrive at the correct fitted model. Key questions should be answered to the best of your ability.

· Should accurately describe and support your model and analysis in words with appropriate amount of detail. Fitted regression equation should be explained in an easy-to-understand way. Any limitations of your analysis should be discussed.)

· Presentation, Organization, Writing

· Overall, the memo should be presented, organized, and written well.