Term Project

7
November 12, 2013 at 6:50 pm  •  Posted in busad265 by  •  7 Comments

Due Thursday, December 5th. 

The Term Project will be an extension of the work you’ve done with the Pueblo Property Data in previous exercises. Here’s what you need to do to complete the project.

1. Produce an appropriate set of graphics and descriptive statistics for your updated neighborhood data. You must use your 30 property records but you can use the other records from your neighborhood as well if you’d like (using more records will probably improve your model accuracy).

2. Build a regression model to estimate Total Value based on one or more independent variables (e.g., Total Finished Square Feet, Number of Bathrooms, etc.). If the best model is bivariate (only one independent variable) that’s fine but be sure to experiment with multiple independent variables and justify your decision regarding what variables are included and excluded. Report the appropriate statistics used to assess how well your model fits the data.

3. Use your model to estimate the value of 5 sample properties. The variables used will depend on Step #2 and the range of values selected is up to you. Just be realistic (e.g., don’t attempt to estimate the value of a 20,000 sq ft home or a home with 16 bathrooms).

4. Use probability theory and statistical inference to determine the accuracy of your 5 sample property estimates. In other words, produce confidence intervals for your estimates.

5. Conduct a Hypothesis Test on your property data. You are responsible for determining the hypothesis you wish to test, setting up the appropriate test, calculating the necessary statistics and interpreting the results.

6. Produce a PowerPoint file to present all steps above. Include slides that summarize your results and conclusions. Your target audience is a typical real estate investor interested in buying single family homes in your assigned neighborhood. What advice would you give the investor?  The Power Point slides should be clear, neat and professional looking. Imagine you’re high paid consultant and you want the investor to pay you for more work and provide good references to other potential clients.

7 Comments

  1. Pingback: Term Project Due This Week | Inferential Statistics and Problem Solving

  2. Anders / December 4, 2013 at 11:40 pm / Reply

    Is it a wrong if we have a negative intercept Coefficients when doing our regression model for when we are making our formula?

    • justinholman / December 4, 2013 at 11:46 pm / Reply

      Not necessarily but it sounds like a strange model with a very steep slope. What independent variables are you using to predict total value?

      • Anders / December 5, 2013 at 12:13 am / Reply

        Total value and first floor area and its giving me a -37567.683 number for my intercept

        • justinholman / December 5, 2013 at 3:54 am / Reply

          Total Value should be the *dependent* variable. Is that the issue?

  3. Anders / December 5, 2013 at 12:16 am / Reply

    That would mean that my equation would be Y=-37567.683+ 123.563x. And that just didnt seem to be right to me.

    • justinholman / December 5, 2013 at 3:54 am / Reply

      I think you might be attempting to predict sq ft using value when you want to be doing the opposite.

Leave a Reply