I have the following regression, where I regress the consumption of gasoline on logarithmic variables.

enter image description here

Meaning of the variables is: pnc=price index for cars pg=price index for gasoline, puc=price index for old cars y=per capita disposable income, ppt=price index for public transportation pd=price index for consumer durables, d79=time dummy for the OPEC oil shock, 1 for the post-1979 period and 0 otherwise

I do not understand how to interpret d79lnpuc variable. Why after the 1979 crisis an increase in the price of old cars has a positive relation to the consumption of gasoline? I think that maybe people started to buy old cars, because of the increase in the price of gasoline. However, I do not see any connection with that and d79lnpuc having a positive coefficient.

I will be thankful for any suggestions and hints.


1 Answer 1


The coefficient for lnpuc is the effect of lnpuc when d79 = 0. This coefficient is -0.324. The coefficient of d79lnpuc measures the change in the effect of lnpus when d79 = 1, thus the effect of lnpuc when d79 = 1 is -0.324 + 0.274 < 0.

  • $\begingroup$ Yes, I understand that. I do not understand why it has a positive coefficient and how to interpret that. $\endgroup$
    – John Doe
    Nov 27, 2021 at 21:00
  • $\begingroup$ It means that when d79 the effect is not as negative compared to when d79 = 0. $\endgroup$ Nov 28, 2021 at 12:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.