Operations Research -Project 1

1. (15 points) With a team of 2 – 3 people, collect 30 – 50 observations (rows):
2. A quantitative dependent (y) variable
3. At least two quantitative independent (x) variables (one of these can be time)
4. At least two categorical independent (x) variables

Describe these variables, including your rationale for choosing them. Your research questions and variable rationale should be non-trivial; in other words, choose your variables for clear, supportable business reasons.

1. (15 points) Analyze and assess the individual variables using MS Excel Descriptive Statistics and appropriate graphs (bar or pie charts; histogram or boxplots).

1. (15 points) Build a multiple regression model to predict y based on your x variables, including scatterplots and normality plot.

1. (40 points) Analyze and assess the model using the tools we discussed; such as strength of slope, interaction variables, R squared, correlation, scatterplots, and/or time series plot, etc. Not all of the tools will be applicable to all of the models, of course. For example, if you collect cross sectional data, you will not make a time series plot.

1. (15 points) Your report on the model should be on a Word document using standard business grammar and usage. Use graphical and tabular output from Excel in your paper to support your assertions about your model. (You do not need to submit a separate Excel document.) Make a final business decision about the utility of the model; would you use this model to make forecasts or predictions?

Extra points are available for originality of topic and depth of analysis. In other words, does the topic choice and rationale reflect a unique business problem? Does the analysis reflect a checklist approach, or does the analysis lead to a creative understanding of the problem?

Please submit one project for your team in blackboard as a word document.

Your project should be of professional quality and should be at a minimum 5-7 pages, double spaced with no greater than 1.25” margins.

Example:

I want to predict Apple’s weekly stock price (quantitative dependent variable) using:

1. Time (quantitative x)
2. The Consumer Price Index (quantitative x)
3. Whether or not Apple released a new product that week (categorical x)
4. Before or after a major iTunes release (categorical x)
5. Week of the month (quantitative or categorical)
6. Before or after the stock price hit \$500 (categorical x)