Cards = number of credit cards used Size = number of individuals in a family Income = family income measured in thousands Autos = number of...

Cards = number of credit cards used

Size = number of individuals in a family

Income = family income measured in thousands

Autos = number of automobiles owned per family

Q1. Use the correlation matrix to determine which independent variable has the strongest relationship with the dependent variable.

Q2. Confirm your results running 3 separate regressions.

Q3. Use your regression of "cards" on "size" to interpret the following: 1) the intercept coefficient, 2) the slope coefficient, and 3) the coefficient of determination. Construct a scattergram that relates predicted values and actual values to "size" (make sure that you "connect" predicted values).

Q4. Now, regress "cards" on both "size" and "income" before reinterpreting the results. Is income statistically significant? Should "income" be retained in the model?

Q5. Regress "cards" on all 3 independent variables. Use Stata to "store" all 3 regressions before constructing a summary table of results. Which model is best? How can we determine whether to add an additional variable to our model?

Q6. Finally, use your final mathematical model to 1) predict card usage for each family in your data set and 2) predict the "errors" for each family in your dataset. Use this data to manually solve for the standard error of the regression (please show your work).

Note: the "id" for the 4th individual should be 4, though it won't affect your analysis!