The following are data on y = quit rate per 100 employees in manufacturing x unemployment rate The data are for the United States and cover the period 1960-1972. Year x Year y x 1960 1.3 6.2 1961 1.2 1962 1.4 5.8 1963 1.4 5.7 1964 1.5 5.0 28870 1967 2.3 3.6 1968 2.5 3.3 1969 2.7 3.3 1970 2.1 5.6 1971 1.8 6.8 1965 1.9 4.0 1972 2.2 5.6 1966 2.6 3.2

(a) Calculate a regression of y on x. y=a+Bx+uobservations or look for other explanations?

(b) Construct a 95% confidence interval for B.

(c) Test the hypothesis Ho: B=0 against the alternative 0 at the 5% significance level.

(d) Construct a 90% confidence interval for = var(u).

(e) What is likely to be wrong with the assumptions of the classical normal linear model in this case? Discuss.