Report

1. Data Description. List of variables, how the data was collected, what kind of experiment it was (observed/designed) and other relevant information.
2. Your goal.
3. Some descriptive statistics. Get summary statistics for all four variables (you can ignore the first variable, that is observation no.) Summary statistics include, mean, sd, five number summary, box-plot and/or histogram.
4. Get a 95% confidence interval for mean sale price, mean land value, mean improvement value, and mean area.
5. Interpret those confidence intervals in non technical terms.
6. The main goal is to find out how the sale price depends on the other three variables (You cannot consider all of them at the same time. That is multiple regression. So, at this point you need do separate analysis for three variables which can be done using simple regression).
7. In particular start with plotting y against x1, y against x2 and y against x3. Get the regression equations to predict the sale price using one of those three variables (one at a time). Report those equations along with MSE and R-square. Which one do you think is the best predictor for sale price. You can determine that by looking at the R-square quantity or by looking at the plots. Bigger R-square means better prediction. See if there are any outliers. Comment.
8. Conclusion: Summarize your findings and explain how this kind of models can be use in real life and their utility.

Price: $20.95

Solution: The downloadable solution consists of 14 pages, 695 words and 8 charts.
Deliverable: Word Document