1. The monthly salary (WAGE), age (AGE), number of years of education beyond the eighth grade (EDUC), and the number of years of experience (EXPER) were obtained for 49 persons in a certain office. The estimated relation between WAGE and the characteristics of a person is as follows (with t-statistics in parentheses): 632.244 + 142.510 EDUC + 43.225 EXPER - 1.913 AGE (3.022) WAGE = (1.493) (4.088) (-0.22) (i) The value of R2 is 0.277. Using this information, test the model for overall significance. (ii) Test the coefficients for EDC and EXPER for statistical significance at the 1% level and AGE for significance at 10% level. (iii) Can you rationalise the negative sign for AGE? (iv) Because the t-statistics is low for AGE, someone suggests that AGE be eliminated. If you follow the suggestion, what kind of specification error might you be making? What are the consequences of that for unbiasedness and consistency unbiasedness and consistency of estimates and forecasts? Explain? (