Skip to content

forecasting accuracy problem in R

3 messages · Jeff Newmiller, Leon

#
Hi,

A few weeks back I used the following command:

accuracy(train,test)

where train and test are training and test data respectively. Last
night I updated R and the forecast package and used the same command
and I got error. After trying a little I used the following command

accuracy(train,test[1:30])

and it worked. (I was checking the accuracy of 30 forecasted values).

Is there some change in the forecast package or did I do something wrong?

regards Leo
#
Please read the posting guide. Packages have their own developers, as well as their own change logs [in this case http://cran.r-project.org/web/packages/forecast/ChangeLog].
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
--------------------------------------------------------------------------- 
Sent from my phone. Please excuse my brevity.
Leo <speedex at gmail.com> wrote:

            
#
Thanks. Could be the following reason given in the changelog.
" accuracy() can now figure out overlapping times for x and f."

best regards
Leo

On Mon, Dec 17, 2012 at 10:11 AM, Jeff Newmiller
<jdnewmil at dcn.davis.ca.us> wrote: