Introductory Econometrics

Answer:

One more year of education is predicted to increase average hourly earnings 8.3%.

Answer:

Many of the datasets do not have information on actual experience and hence researchers use potential experience as a proxy for actual experience. Potential variable is especially likely to overstate actual experience for women because of the amount of time that women spend out of the work force. Also the amount that potential experience overstates actual experience varies systematically with other variables, such as race and education, possibly leading to the biased estimate of the coefficients on these other variables.

Answer:

If age=40, education=12, Exper=40-12-6=22

If age=60, education=12, Exper=60-12-6=42

So if experience increases from 22 to 23 the value of ln(Edarn) changes from

ln(Edarn)=-0.01+0.101*12+0.033*22-0.0005*22*22=1.686

ln(Edarn)=-0.01+0.101*12+0.033*23-0.0005*23*23=1.697

so that education on earnings changes from

Edarn=5.3978

Edarn=5.4548

So % change in education on earnings changes is in the increasing direction and is=(5.4548-5.3978)*100/5.3978

=1.0555%

So if experience increases from 42 to 43 the value of ln(Edarn) changes from

ln(Edarn)=-0.01+0.101*12+0.033*42-0.0005*42*42=1.706

ln(Edarn)=-0.01+0.101*12+0.033*43-0.0005*43*43=1.6965

so that education on earnings changes from

Edarn=5.5069

Edarn=5.4548

So % change in education on earnings changes is in the decreasing direction and is =(1.706-1.6965)*100/1.706

=0.557%

Answer:

T(edu)=0.101/0.012=8.417

T(exper)=0.033/0.006=5.5

T(exper2)=-0.0005/0.0001=-5

T(0.05, n-2)=t(0.05, 1523-2)=t(0.05, 1521)=1.962

As the absolute values of each t-statistic are greater than 1.962, we conclude that all the coefficients are statistically significant.

The coefficient on education changed so little because some of the effect of education is depicted by the experience variable as

Experience=age-education-6...