I am trying to use the "car" command in "cts package" in R program and I see the "scale" parameter there. I wonder whether this can be assumed to be equivalent to time intervals for time series forecasting. For example, the code is like this:

car(x, y=NULL, scale = 1.5, order = 3, ctrl=car_control())

and the official cts package explanation is the following:

scale: The kappa value referred to in the paper by Belcher et a. (1994).We now recommend selection of kappa along with the model order by using AIC. Also, it is suggested to choose kappa close to 2pi times 1/mean.delta (reciprocal of the mean time between observations), though it is a good idea to explore somewhat lower and higher values to see whether the spectrum estimates were sensitive to this choice. Choosing kappa lower increases the risk of trying to estimate the spectrum beyond the effective Nyquist frequency of the data - though this does depend on the distribution of intersample times.

Can anyone have some ideas, please..?


1 Answer 1


It says explicitly in the section you quote:

"it is suggested to choose $κ$ close to $2π$ times [... the] reciprocal of the mean time between observations".

So it's proportional to frequency of observations (1/average time interval) not the time interval. [They also give a reference which no doubt has an explicit definition as well.]


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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