Suppose I have a data with Gaussian copula with ($\rho =0.8$). Suppose further the contourHere plot is as follows:

My question is:

How can I interpret the result with respect to my original values? For example, suppose that the first variable is the temperature and the second one is the bike rental. Suppose that I would like to explain at what values these variables become independent, from the plot. For example, I can say that these two variables are positively correlated at the interval $[0.4,0.6]$ (for example).

Now I would like to know if there is a way that I can say what is the corresponding temperature? That is, what is the corresponding temperature for $[0.4,0.6]$?

  • $\begingroup$ You don't have the information to do this. You need the maps from your original variables to the $(u_1,u_2)$ coordinates in the image. $\endgroup$ – whuber Jun 30 at 13:27
  • $\begingroup$ @whuber Thank you so much for your comment. Is there a way to do so? Thanks. $\endgroup$ – Maryam Jun 30 at 13:32
  • $\begingroup$ Yes: review the process you used to produce the copula. The initial step must construct those two maps. There's no other way. $\endgroup$ – whuber Jun 30 at 13:36
  • $\begingroup$ @whuber thank you again. I really appreciate your help. However, I really do not get your comment. I just use pobs to get copula data and then I plot the contour. If I plot my original data then, it is not helpful. $\endgroup$ – Maryam Jun 30 at 13:58
  • $\begingroup$ You will need to consult your software documentation to learn how to extract this information. $\endgroup$ – whuber Jun 30 at 14:16

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.