prepare a statistical analysis of single-family home values currently in three zip codes in West County St. Louis.
Part I Descriptive Statistics (30%)
You need to prepare a workbook that summarizes and charts data you feel is relevant to understanding the characteristics ofsingle family residential homesin this area overall and differences between zip codes. This is an intentionally open-ended assignment similar to what you might receive in industry. Use your descriptive statistics chapters for ideas and label your worksheets D1, D2, . . . etc and use text boxes to describe the contents and conclusions for each sheet. Your sheets should summarize a single characteristic and use any descriptive statistics you feel useful in presenting the data including summary measures, pivot table reports, and charts. A minimum of three characteristics (three sheets) is required and you can have multiple items on each sheet summarizing that characteristic. Select the most important characteristics you feel like list price and other characteristics that might impact the value of the home and make sure at least one of the characteristics is a categorical (qualitative) data item. Format the sheets and contents appropriately and consistently to communicate your summary to your audience. You are summarizing, so avoid too much information.
The data has not been "cleansed" so it may be necessary to remove observations (rows) that seem inappropriate for your summary. Document any data elements item you removed and why by cutting and pasting them below the original data table and adding a comment after the item summarizing your justification for removing the data.
Part 2 Inferential Statistics (20%) - (the items in parenthesis are examples, you do NOT have to do that particular analysis)
Create a worksheet called I1 in which you create a fewconfidence intervals for the mean values of quantitative characteristics (e.g. average list price in each zip code).
Create a worksheet called I2 in which you create a number of confidence intervals for the proportion of a categorical (qualitative) characteristic (e.g. % of homes with four bedrooms or more).
Create a worksheet called I3 in which you create hypothesis tests of two means appropriate for this data (e.g. determine if the average price by location is different in two zip codes).
Create a worksheet called I4 in which you create hypothesis tests of more than two means (Single-Factor ANOVA) appropriate for this data (e.g determine if the average price if different in different school districts).
Part 3 Regression Model (40%)
Create a regression model to determine the best (based on R-squared)model to help predict list price (Y) containingone significant independent variable (X) in worksheet labeled M1.
Create a regression model to determine the best(based on R-squared) model to help predict list price (Y) containingtwo significant independent variables (X) in worksheet labeled M2.
Part 4 Optional: For A consideration (10%)
Create the bestregression model to help predict list price for all or a subset of the data(based on R-squared) model containingmore than twosignificant independent variablesin worksheet MX. Summarize your model and conclusions in a text box and use your regression model to predict the value of an example home.
Keep a copy of all intermediate models attempted and the reasons why variables were added or removed from the model and include a sheet called Log that summarizes each model you tried and what you did in that model.