[R] Difference between spec.pgram & spec.ar

[Prof Brian Ripley]

On 09/04/2012 18:52, Bert Gunter wrote:
> On Mon, Apr 9, 2012 at 9:27 AM, Bazman76<h_a_patience at hotmail.com>  wrote:
>> Yes I agree, there may be something pathalogical in the way at least one of
>> the models handles the data.  That's why I was trying to get a better handle
>> on how the two functions spec.prgm() and spec.ar() work.
>>
>> The data has been processed by a wavelet analysis, so what you are seeing as
>> the "raw" data is in fact the level1 details from the wavelet anlaysis.
>>
>> In theory this should only have high frequency components, that was why I am
>> so surpirsed to see such strong components at low frequencies.
>>
>> That is not a R quesiton per se, but surely how spec.prgm() and spec.ar()
>> work is?
>
> Not necessarily, if they are e.g. C or Fortran programs merely called
> by R. Indeed, even if written in R, if the algorithms are the issue,
> then that is essentially a statistics/numerical analysis matter, not
> an R programming one.

Which is why the help pages often (and do here) have definitive 
references.  So the best course of action is to start following up the 
references.

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595