[Met_help] [rt.rap.ucar.edu #68233] History for query reg the R scripts on your website

[John Halley Gotway via RT]

----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

Dear John, 
this is a query reg the R scripts on your website
I saw this PDF file available at 
http://www.dtcenter.org/met/users/downloads/Rscripts/mpr_plots.pdf

It has many plots. It seems to me that the Figs on Pg 1 and 2 are the same.  Many Figs I feel are the same. 


secondly, the R script that you have on yor website for making these diagrams, 
for running this script, what I should have on my AIX system.

Or can I run this script on windows too. as it requres the MPR o/p files only. 




geeta 		 	   		  

----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: query reg the R scripts on your website
From: John Halley Gotway
Time: Thu Jul 17 10:15:07 2014

Geeta,

That plot_mpr.R Rscript (
http://www.dtcenter.org/met/users/downloads/Rscripts/plot_mpr.R) is
provided as a sample showing how one can read MET output into R and
perform
additional analysis.  It is set up to read METv3.0 MPR lines from
either
the "_mpr.txt" files or the ".stat" files.  It looks at the header
information in the data and picks out the set of unique cases, defined
as
the combination of the model name, forecast variable/level,
observation
variable/level, observation type, masking region, and interpolation
method.

For each unique case, it collects the group of matched pairs for that
case
and creates 4 plots on each page of the PDF output file:
 - histogram of the forecast values (top-left panel)
 - histogram of the observation value (top-right panel)
 - scatter plot of the forecast and observation pairs (lower-left
panel)
 - and a quantile-quantile plot (lower-right panel)

If you look at the plot titles, you'll see that the plots on the first
page
of that PDF file differ slightly from those on the second page.  The
data
is almost the same, but the first page is for the nearest neighbor
interpolation method (UW_MEAN(1)) while the second page is for the
median
of the 9 closest points (MEDIAN(9)).

I don't have access to an AIX machine, but as long as R and Rscript
are
installed on that machine, you should be able to run this script.  You
should also be able to run this on a windows machine.

Please feel free to modify that script in any way you need to perform
whatever type of additional analysis you'd like to do.  Like I said,
it's
meant as an example to help users get going.

Thanks,
John Halley Gotway
met_help at ucar.edu


On Thu, Jul 17, 2014 at 8:22 AM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:

>
> Thu Jul 17 08:22:43 2014: Request 68233 was acted upon.
> Transaction: Ticket created by geeta124 at hotmail.com
>        Queue: met_help
>      Subject: query reg the R scripts on your website
>        Owner: Nobody
>   Requestors: geeta124 at hotmail.com
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68233 >
>
>
> Dear John,
> this is a query reg the R scripts on your website
> I saw this PDF file available at
> http://www.dtcenter.org/met/users/downloads/Rscripts/mpr_plots.pdf
>
> It has many plots. It seems to me that the Figs on Pg 1 and 2 are
the
> same.  Many Figs I feel are the same.
>
>
> secondly, the R script that you have on yor website for making these
> diagrams,
> for running this script, what I should have on my AIX system.
>
> Or can I run this script on windows too. as it requres the MPR o/p
files
> only.
>
>
>
>
> geeta
>

------------------------------------------------