[R] Problems with lambda argument

Paul Bernal p@u|bern@|07 @end|ng |rom gm@||@com
Mon Nov 18 15:30:37 CET 2019


Dear friend Duncan,

Thank you for your kind reply, I appreciate the fact that you always try to
help. I got the error fixed, to fix the problem, instead of setting
lambda="auto", I did lambda=BoxCox.lambda(tsobject), not sure if this is
equivalent to setting lambda="auto", but I guess is a workaround at the
very least.

Cheers,

Paul

El lun., 18 nov. 2019 a las 9:26, Duncan Murdoch (<murdoch.duncan using gmail.com>)
escribió:

> On 18/11/2019 8:44 a.m., Paul Bernal wrote:
> > Dear friends,
> >
> > Hope you are doing great. When setting the lambda = "auto" in the
> > auto.arima function, I get an error message saying non-numeric argument
> to
> > binary operator.
> >
> > I have performed several tests changing parameters and going through the
> > other lines of code and as soon as I set lambda = "auto" the error pops
> up,
> > otherwise the model runs perfectly fine.
> >
> > Any guidance will be greatly appreciated,
>
> If I run the example(auto.arima) code from the forecast package but add
> lambda="auto", I don't get any error, so I think you'll have to post a
> reproducible example if you want help.
>
> Duncan Murdoch
>

	[[alternative HTML version deleted]]



More information about the R-help mailing list