Skip to content

Any Unsupervised Learning Algorithm for Time Series Forecasting in R

4 messages · Paul Bernal, Bert Gunter, Hasan Diwan

#
Dear friends,

Hope you are all doing great. I would like to know if R has any
unsupervised algorithm to generate forecasts for historical data.

Any help will be greatly appreciated,

Best regards,

Paul
#
Depending on exactly what you mean by"unsupervised", many.

See here under "Decomposition and filtering":

https://cran.r-project.org/web/views/TimeSeries.html

You could also search on something like "smooth time series R" etc.

However, assuming I have correcty interpreted "unsupervised algorithm," be
aware that extrapolating from such smoothed historical data can be
problematic precisely because no structure/model has been specified. See
here for some further background and R functions:

http://a-little-book-of-r-for-time-series.readthedocs.io/en/latest/src/timeseries.html#holt-winters-exponential-smoothing


Cheers,
Bert






Bert Gunter

"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Wed, Jun 20, 2018 at 9:00 AM, Paul Bernal <paulbernal07 at gmail.com> wrote:

            

  
  
#
... and II should have added that the "forecast" package may be what you're
looking for.

Cheers,
Bert



Bert Gunter

"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Wed, Jun 20, 2018 at 9:30 AM, Bert Gunter <bgunter.4567 at gmail.com> wrote:

            

  
  
#
Paul,
On Wed, 20 Jun 2018 at 09:04, Paul Bernal <paulbernal07 at gmail.com> wrote:
Yes , it does. Perhaps you'd be kind enough to provide a sample of
your data --dput(sample(pauls.data)) on gist.github.com -- and what
you wish to achieve with it? Many thanks! -- H