I'd like to know if it's correct to the CV function in the forecast R package (http://cran.r-project.org/web/packages/forecast/forecast.pdf) to cross-validate a nonlinear model that is linear in the parameters. The description in the documentation says that the CV function "Computes the leave-one-out cross-validation statistic (also known as PRESS – prediction residual sum of squares), AIC, corrected AIC, BIC and adjusted R^2 values for a linear model."


  • $\begingroup$ What do you mean by a "nonlinear model that is linear in the parameters"? Do you mean to say that it is nonlinear in the variables? $\endgroup$
    – abaumann
    Commented Oct 3, 2014 at 14:03
  • $\begingroup$ Yes, nonlinear in the variables, but linear in the parameters. $\endgroup$ Commented Oct 4, 2014 at 17:19

1 Answer 1


If a model is linear in parameters, we'd normally call it a linear model.

(It's also linear in transformed variables, the ones actually supplied to a regression function.)

So yes, it would be correct to use that CV function, since it's talking about being for the model you have.


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.