Suppose I have a time-series data (temperature and ice-cream sales). I would like to fit a copula model to this data (just to understand the dependency between these two variables. I do not want to fit a regression copula model). My question is, do I need to detrend the data? As I understand I can detrend the data only if I need to fit a regression model.
Asked
Active
Viewed 51 times
2
-
You don't need to de-trend if you just want to fit a regression model, at least not in general. Are you fitting a linear model? – Adrian Keister Jul 13 '20 at 18:23
-
@AdrianKeister I try to fit a copula model – Maryam Jul 13 '20 at 20:25