Use the data in MINWAGE.RAW for this exercise, focusing on the wage and employment series for sector 232 (Men's and Boys' Furnishings). The variable gwagelil is the monthly growth (change in logs) in the average wage in sector 232; gempTil is the growth in employment in sector 232; gmwage is the growth in the federal minimum wage; and gcpi is the growth in the (urban) Consumer Price Index,
(i) Find the first order autocorrelation in gwage232. Does this series appear to be weakly dependent?
(ii) Estimate the dynamic model
gwage232, = (0 + (1 gwage232t-1 + (2 gmwage, + (3 gcpi, + u,
by OLS. Holding fixed last month's growth in wage and the growth in the CPI, does an increase in the federal minimum result in a contemporaneous increase in gwage232t? Explain.
(iii) Now add the lagged growth in employment, gemp232t-1 to the equation in part (ii). Is it statistically significant?
(iv) Compared with the model without gwage232t-1 and gemp232t-1, does adding the two lagged variables have much of an effect on the gmwage coefficient?
(v) Run the regression of gmwaget, on gwage232t-1 and gemp232t-l, and report the R-squared. Comment on how the value of R-squared helps explain your answer to part (iv).