Consider the following model: In WAGE_i = beta_0 + beta_1 EDUC_i + beta_2 EXPER
ID: 3201737 • Letter: C
Question
Consider the following model: In WAGE_i = beta_0 + beta_1 EDUC_i + beta_2 EXPER + beta_3 EXPER^2 _i + u_i where In WAGE denotes the logarithm of wages of married women, EDUC is years of education, and EXPER is years of experience. The model was estimated using OLS (standard errors in parentheses, n = 750): In WAGE_i =-0.5220_(0.1980) + 0.1075 EDUCi_(0.0141) +0.0416 EXPER_(0.0132) -0.0008 Expert (0.0004) Using estimated coefficients, describe the impact of changes in education and experience on wages. Test each regression coefficient for significance at the 1% significance level and report corresponding p-values. Test at the 5% level the hypothesis that the return to education exceeds 0.1.Explanation / Answer
(a) since the coefficient of Education and experience is positive so its impact of change is positive. as these variable will increase the wage also increases and vice versa
(b) we use t-test for test of significance and t=(regression coefficeint)/SE(regression coefficeint)
using the ms-excel command =TDIST(0.522,746,2) the p-value for slope is given by 0.0087 and similarly for other coefficeints.
since p-value is less than 0.01 for slope, educ and exper so thses variable is significant at 1% significance level
but exper2 is not significant as its p-value is more than 0.01
(c) here we use t-test and t=(0.1075-0.1)/0.0141= 0.5319 is less than
the its one tailed critical value=1.6469 at 5% level of significance which means , return to education exceeds 0.1 is not significant.
variable coefficeint SE t p-value slope -0.522 0.1986 -2.6284 0.00875525 education 0.1075 0.0141 7.624113 7.48531E-14 exper 0.0416 0.0132 3.151515 0.001689195 Exper2 -0.0008 0.0004 -2 0.045862355