Which Regression Model Is More Realistic: Linear or Logarithmic?

[wow007051]

there are two regreesion model Eviews output... which one is more realistic?

model 1:
wagehat = 116.9916 + 8.303*IQ
model 2:
logwagehat = 5.88 + 0.0088*IQ

both of them use same samples...

do i need to know futher statistic information to judge which one is more realistic? such as Rsquare or some thing?

 

[Ray Vickson]

there are two regreesion model Eviews output... which one is more realistic?

model 1:
wagehat = 116.9916 + 8.303*IQ
model 2:
logwagehat = 5.88 + 0.0088*IQ

both of them use same samples...

do i need to know futher statistic information to judge which one is more realistic? such as Rsquare or some thing?

This question is impossible to answer without knowing the actual data. You can find more info in http://en.wikipedia.org/wiki/Regression_model_validation . Click on the external link "How can I tell if a model fits my data" to see a lot more on this topic.

RGV