0
$\begingroup$

I want to generate two artificial time series, one of which acts as the explanatory variable and the other as the dependent variable for a regression model. Can anybody suggest as to how to proceed with this?

$\endgroup$
1
$\begingroup$

Multiply the Y dataset by .7 and use that outcome as the causal. Of course, you want to add in some noise. You may want to complicate things by adding in an outlier(yes, they exist in causals).

$\endgroup$
  • $\begingroup$ Thanks for the quick reply. However, I don't get how rescaling helps; could you elaborate please? $\endgroup$ – Keshaw Jun 21 '16 at 13:12
  • $\begingroup$ No rescaling being done here. I was trying create data that would follow a model like this Y =u +.7x $\endgroup$ – Tom Reilly Jun 21 '16 at 13:29
  • $\begingroup$ Basically @Tom Reilly has suggested (1) build the relationship model first then (2)a created time series of the independent variable then (3) calculate the dependent variable time series. $\endgroup$ – MarkR Jun 21 '16 at 20:49
  • $\begingroup$ Thanks for clearing this thing up. But actually I want to simulate datasets to check validity of my regression model $\endgroup$ – Keshaw Jun 22 '16 at 5:24

Your Answer

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

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