MSA STAT Workshop Assessment

REQUIREMENTS

1. What is the population Regression function? ("GNP"  is dependent, and "Invest," "CPI" and "Interest" are independent)

2. Is this regression function significant overall (state the null and alternative hypotheses and explain how you reached that conclusion)?

3. Interpret the coefficient associated with "Invest."

4. Which independent variables offer any predictability of the dependent one (regardless of whether some of the variation is shared with other variables)?  Explain.

5. Which independent variables offer any unique predictability of the dependent one? Explain.

6. How would your answers to the previous question change if you were testing significance at an α level of 10% instead of 5%?

7. What proportion of the variation in the dependent variable is predictable from the set of independent variables in the sample?

8. What is the expected defense budget for a year where "Invest"= 411.7; "CPI"= 211.54; and "Interest"= 8.48?  (For the sake of this exercise, do not exclude the insignificant variables from your calculation)

9. Test your model for Multicollinearity using the VIF Method.

10. Test your model for Heteroskedasticity using the White's Test.

11. Test your model for Autocorrelation using the Durbin-Watson Test.

12. Do you have any reason to believe that GNP is trending?  That is, increasing or decreasing at some significant average rate per year.  (hint: you need to test this using some regression)

DATASET -

The dataset TableF3-1.txt contains 15 Yearly Observations (1968-1982) on the following variables:

Year = Date,

GNP = Nominal GNP,

Invest = Nominal Investment,

CPI = Consumer price index,

Interest = Interest rate (average yearly discount rate at the New York Federal Reserve Bank)

Source: Economic Report of the President.

Attachment:- Assignment.rar