I read Discrete returns versus log returns of assets and it is very helpful.

However, if log-returns are easier for time-aggregation, then why do economists work with discrete returns e.g. in GDP growth?

  • $\begingroup$ Because in the GDP case, it doesn't make sense to subdivide your time into intervals much shorter than say a quarter. Unless, you look at very long time periods, this doesn't approximate well by a continuous scheme. $\endgroup$ – Raskolnikov Oct 31 '17 at 7:50
  • $\begingroup$ @Raskolnikov: well that makes it useful for very long time periods I would say ? Shouldn't the sum of quarterly growths equal annual growth ? And let's say that GDP goes from 100 in year 1 to 110 in year 2 and back to 100 in year 3 then growth from year 1 to year 2 is +10% while from year 2 to year 3 it is -9.1%, would it be ''nicer'' if these growth numbers were ''symmetric'' ? $\endgroup$ – user0009 Oct 31 '17 at 9:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.