[Met_help] [rt.rap.ucar.edu #82019] History for mctc.tx output

[John Halley Gotway via RT]

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

When I was running grid_stat, I had the following thresholds defined in my
config file.

cat_thresh = [ >=0.0, >=5.0 , >=10.0];

In the output of the mctc.txt file I was expecting a 3x3 contingency table
yet the N_CAT was 4 with 16 F_O values (see attached file).

Can you offer an explanation why N_CAT is 4 in this case, its likely my
misunderstanding but need clarification.

Thanks, Brad DIehl


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

Subject: mctc.tx output
From: John Halley Gotway
Time: Fri Sep 15 12:48:21 2017

Brad,

I see you have a question about how thresholds translate into
multi-category contingency tables.

If you think about it, 1 threshold is used to define a 2x2 contingency
table.  So 1 threshold means 2 categories.

Likewise, 3 thresholds defines a 4x4 table.

In your case, values could either be:
(1) less than 0
(2) between 0 and 5
(3) between 5 and 10
(4) greater than 10

I'm guessing that you're processing precip which can't actually be
less
than 0.  For a 3x3 table, use 2 thresholds:

cat_thresh = [ >=5.0 , >=10.0];

Thanks,
John



On Fri, Sep 15, 2017 at 12:21 PM, Brad Diehl - NOAA Federal via RT <
met_help at ucar.edu> wrote:

>
> Fri Sep 15 12:21:55 2017: Request 82019 was acted upon.
> Transaction: Ticket created by brad.diehl at noaa.gov
>        Queue: met_help
>      Subject: mctc.tx output
>        Owner: Nobody
>   Requestors: brad.diehl at noaa.gov
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=82019 >
>
>
> When I was running grid_stat, I had the following thresholds defined
in my
> config file.
>
> cat_thresh = [ >=0.0, >=5.0 , >=10.0];
>
> In the output of the mctc.txt file I was expecting a 3x3 contingency
table
> yet the N_CAT was 4 with 16 F_O values (see attached file).
>
> Can you offer an explanation why N_CAT is 4 in this case, its likely
my
> misunderstanding but need clarification.
>
> Thanks, Brad DIehl
>
>

------------------------------------------------
Subject: mctc.tx output
From: Brad Diehl - NOAA Federal
Time: Fri Sep 15 13:04:25 2017

John,

Once again, my misunderstanding and your clarifying it, now I see how
you're using the thresholds from your example. It's not that I wanted
a 3x3
table, its just from my misunderstanding I was expecting one. From
your
example I see how the binning goes. You can close this out. See you
next
week at NCEP.

On Fri, Sep 15, 2017 at 6:48 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:

> Brad,
>
> I see you have a question about how thresholds translate into
> multi-category contingency tables.
>
> If you think about it, 1 threshold is used to define a 2x2
contingency
> table.  So 1 threshold means 2 categories.
>
> Likewise, 3 thresholds defines a 4x4 table.
>
> In your case, values could either be:
> (1) less than 0
> (2) between 0 and 5
> (3) between 5 and 10
> (4) greater than 10
>
> I'm guessing that you're processing precip which can't actually be
less
> than 0.  For a 3x3 table, use 2 thresholds:
>
> cat_thresh = [ >=5.0 , >=10.0];
>
> Thanks,
> John
>
>
>
> On Fri, Sep 15, 2017 at 12:21 PM, Brad Diehl - NOAA Federal via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Fri Sep 15 12:21:55 2017: Request 82019 was acted upon.
> > Transaction: Ticket created by brad.diehl at noaa.gov
> >        Queue: met_help
> >      Subject: mctc.tx output
> >        Owner: Nobody
> >   Requestors: brad.diehl at noaa.gov
> >       Status: new
> >  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=82019 >
> >
> >
> > When I was running grid_stat, I had the following thresholds
defined in
> my
> > config file.
> >
> > cat_thresh = [ >=0.0, >=5.0 , >=10.0];
> >
> > In the output of the mctc.txt file I was expecting a 3x3
contingency
> table
> > yet the N_CAT was 4 with 16 F_O values (see attached file).
> >
> > Can you offer an explanation why N_CAT is 4 in this case, its
likely my
> > misunderstanding but need clarification.
> >
> > Thanks, Brad DIehl
> >
> >
>
>

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