Please read these directions carefully.

This semester you will be required to do one project broken down into 3 parts. This part uses the data you gathered in Part 1.

For this part, please do the following:

For this project, use your data and your two quantitative variables to estimate the best fitting line through your data, using price as the dependent variable, y, andmileage as the independent variable, x.

Using StatCrunch perform a linear regression analysis.

Your project should be done in a semi-professional looking report, preferably typed in Word with all the parts below.

1. Introduction: Briefly describe your sample and your two quantitative variables, including their units of measurement. State that the purpose of your analysis is to use linear regression to predict price based on mileage. - 1 points
2. Insert a scatter plot with mileage on the x axis and price on the y-axis. Conduct simple linear regression, with mileage as the x (independent) variable and price as the y (dependent) variable. Copy/paste your scatterplot and regression results into the report. - 2 points
3. Find & practically interpret the correlation coefficient (r) - 4 points
4. Find & practically interpret the coefficient of determination (R²) - 4 points
5. Find the equation of the least-squares line. Specifically, fill in the numbers for "a" and "b" in the least-squares equation: y^=a+bx When fitting this model, have y be price and x be mileage. - 4 points
6. Find & practically interpret the slope (b) - 4 points
7. Find & interpret the y-intercept (a) if an interpretation is possible. If it's not possible, just report the y-intercept and explain whyit can not be interpreted practically- 2 points
8. Use your model to estimate the price of a car that has 100,000 miles (it's required to include this prediction in the report, whether or not it's good practice to make this prediction). Is this prediction reasonable? Why or why not? - 3 point
9. Provide a brief conclusion or discussion of the results that includes one of the following: - 1 point

• • potential problems or sources of bias
  • are the results what you expected?
  • are there other variables that may also be useful to predict the price of a car