In my case non-overlapping data would represent the scenario where futures prices (3 months) do not correspond to the futures spot prices in terms of delivery date. For example, futures settlement price (with 3 months delivery) on 01/01/2016 would represent the futures spot price on 01/04/2016 or should I just make 84 days horizon? Since various months consistent of different amount of days.

Chowdhury, A. R. (1991), Futures market efficiency: Evidence from cointegration tests. J. Fut. Mark., 11: 577–589. doi: 10.1002/fut.3990110506

I am trying to replicate this study, btw. So how exactly I should work out non-overlapping data?

  • 1
    $\begingroup$ It is hard to suggest a general technique, but in linear regression you should use Newey-West correction. $\endgroup$ – onlyvix.blogspot.com Feb 24 '16 at 19:02
  • $\begingroup$ If you have downloaded the future price data, maturity date is always given. By the way, if you have problem with overlapping data then go through this paper: www2.warwick.ac.uk/fac/soc/wbs/subjects/finance/faculty1/… $\endgroup$ – Neeraj Feb 25 '16 at 18:08

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.