[Met_help] [rt.rap.ucar.edu #56643] History for error in fho file txt of PointStat
Paul Oldenburg via RT
met_help at ucar.edu
Mon Jun 11 09:04:11 MDT 2012
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Dear all,
I want to report an error in the FHO file output of PointStat Tool in MET3.1. The FHO txt file has in the 23 column the F_RATE, in the 24 the H_RATE and in the 25 the O_RATE. I suppose that the calculation of the H_RATE is not correct, in fact, if I control the n values in the contingency table from the CTC text file, I find that H_RATE in the FHO text file is equal to n11/T (T=n11+n10+n01+n00) instead the correct H_RATE should be equal to n11/(n11+n01). In the output file, in fact, I find H_RATE values very low due to this error.
I hope my suggestion will be helpful.
Best regards
Sabrina
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Sabrina Gentile, Ph.D
CETEMPS, University of L'Aquila
Department of Physics - Atmospheric Sciences
Via Vetoio, 67010 Coppito L'Aquila
Italy
E-mail: sabrina.gentile at aquila.infn.it
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #56643] error in fho file txt of PointStat
From: Paul Oldenburg
Time: Tue May 29 11:12:47 2012
Sabrina,
Thank you for reporting this problem. It appears that you are correct
that the H_RATE calculation in the FHO line is
incorrect. There is an equivalent calculation called PODY in the CTC
line, which appears to be calculated correctly.
Please refer to this quantity until we are able to deploy a bug fix
for the error in the FHO H_RATE calculation. I will
let you know when a bug fix is available for this problem.
Thanks,
Paul
On 05/28/2012 07:52 AM, Sabrina Gentile via RT wrote:
>
> Mon May 28 07:52:20 2012: Request 56643 was acted upon.
> Transaction: Ticket created by sabrina.gentile at aquila.infn.it
> Queue: met_help
> Subject: error in fho file txt of PointStat
> Owner: Nobody
> Requestors: sabrina.gentile at aquila.infn.it
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=56643>
>
>
> Dear all,
> I want to report an error in the FHO file output of PointStat Tool
in MET3.1. The FHO txt file has in the 23 column the F_RATE, in the 24
the H_RATE and in the 25 the O_RATE. I suppose that the calculation of
the H_RATE is not correct, in fact, if I control the n values in the
contingency table from the CTC text file, I find that H_RATE in the
FHO text file is equal to n11/T (T=n11+n10+n01+n00) instead the
correct H_RATE should be equal to n11/(n11+n01). In the output file,
in fact, I find H_RATE values very low due to this error.
> I hope my suggestion will be helpful.
> Best regards
>
>
>
>
> Sabrina
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Sabrina Gentile, Ph.D
> CETEMPS, University of L'Aquila
> Department of Physics - Atmospheric Sciences
> Via Vetoio, 67010 Coppito L'Aquila
> Italy
> E-mail: sabrina.gentile at aquila.infn.it
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #56643] error in fho file txt of PointStat
From: Paul Oldenburg
Time: Tue May 29 12:19:09 2012
Sabrina,
A new patch for METv3.1 has been deployed to the MET website. You can
download the patch here:
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php.
The patched code should correctly calculate
the FHO H_RATE statistic. Thanks again for bringing this matter to
our attention. If you have any questions or notice
any other problems, please let us know.
Thanks,
Paul
On 05/29/2012 11:12 AM, Paul Oldenburg wrote:
> Sabrina,
>
> Thank you for reporting this problem. It appears that you are
correct that the H_RATE calculation in the FHO line is
> incorrect. There is an equivalent calculation called PODY in the CTC
line, which appears to be calculated correctly.
> Please refer to this quantity until we are able to deploy a bug fix
for the error in the FHO H_RATE calculation. I will
> let you know when a bug fix is available for this problem.
>
> Thanks,
>
> Paul
>
>
> On 05/28/2012 07:52 AM, Sabrina Gentile via RT wrote:
>>
>> Mon May 28 07:52:20 2012: Request 56643 was acted upon.
>> Transaction: Ticket created by sabrina.gentile at aquila.infn.it
>> Queue: met_help
>> Subject: error in fho file txt of PointStat
>> Owner: Nobody
>> Requestors: sabrina.gentile at aquila.infn.it
>> Status: new
>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=56643>
>>
>>
>> Dear all,
>> I want to report an error in the FHO file output of PointStat Tool
in MET3.1. The FHO txt file has in the 23 column
>> the F_RATE, in the 24 the H_RATE and in the 25 the O_RATE. I
suppose that the calculation of the H_RATE is not
>> correct, in fact, if I control the n values in the contingency
table from the CTC text file, I find that H_RATE in the
>> FHO text file is equal to n11/T (T=n11+n10+n01+n00) instead the
correct H_RATE should be equal to n11/(n11+n01). In
>> the output file, in fact, I find H_RATE values very low due to this
error.
>> I hope my suggestion will be helpful.
>> Best regards
>>
>>
>>
>>
>> Sabrina
>>
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> Sabrina Gentile, Ph.D
>> CETEMPS, University of L'Aquila
>> Department of Physics - Atmospheric Sciences
>> Via Vetoio, 67010 Coppito L'Aquila
>> Italy
>> E-mail: sabrina.gentile at aquila.infn.it
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #56643] error in fho file txt of PointStat
From: Sabrina Gentile
Time: Tue Jun 05 02:51:05 2012
Dear Paul,
I downloaded and reinstalled the patche
METv3.1_patches_20120501.tar.gz from
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
but
I found again the previous problem in the FHO H_RATE statistic.
Thanks, bye
Sabrina
----- Original Message -----
From: "Paul Oldenburg via RT" <met_help at ucar.edu>
To: <sabrina.gentile at aquila.infn.it>
Sent: Tuesday, May 29, 2012 8:19 PM
Subject: Re: [rt.rap.ucar.edu #56643] error in fho file txt of
PointStat
> Sabrina,
>
> A new patch for METv3.1 has been deployed to the MET website. You
can
> download the patch here:
>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php.
> The patched code should correctly calculate
> the FHO H_RATE statistic. Thanks again for bringing this matter to
our
> attention. If you have any questions or notice
> any other problems, please let us know.
>
> Thanks,
>
> Paul
>
>
> On 05/29/2012 11:12 AM, Paul Oldenburg wrote:
>> Sabrina,
>>
>> Thank you for reporting this problem. It appears that you are
correct
>> that the H_RATE calculation in the FHO line is
>> incorrect. There is an equivalent calculation called PODY in the
CTC
>> line, which appears to be calculated correctly.
>> Please refer to this quantity until we are able to deploy a bug fix
for
>> the error in the FHO H_RATE calculation. I will
>> let you know when a bug fix is available for this problem.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 05/28/2012 07:52 AM, Sabrina Gentile via RT wrote:
>>>
>>> Mon May 28 07:52:20 2012: Request 56643 was acted upon.
>>> Transaction: Ticket created by sabrina.gentile at aquila.infn.it
>>> Queue: met_help
>>> Subject: error in fho file txt of PointStat
>>> Owner: Nobody
>>> Requestors: sabrina.gentile at aquila.infn.it
>>> Status: new
>>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=56643>
>>>
>>>
>>> Dear all,
>>> I want to report an error in the FHO file output of PointStat Tool
in
>>> MET3.1. The FHO txt file has in the 23 column
>>> the F_RATE, in the 24 the H_RATE and in the 25 the O_RATE. I
suppose
>>> that the calculation of the H_RATE is not
>>> correct, in fact, if I control the n values in the contingency
table
>>> from the CTC text file, I find that H_RATE in the
>>> FHO text file is equal to n11/T (T=n11+n10+n01+n00) instead the
correct
>>> H_RATE should be equal to n11/(n11+n01). In
>>> the output file, in fact, I find H_RATE values very low due to
this
>>> error.
>>> I hope my suggestion will be helpful.
>>> Best regards
>>>
>>>
>>>
>>>
>>> Sabrina
>>>
>>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> Sabrina Gentile, Ph.D
>>> CETEMPS, University of L'Aquila
>>> Department of Physics - Atmospheric Sciences
>>> Via Vetoio, 67010 Coppito L'Aquila
>>> Italy
>>> E-mail: sabrina.gentile at aquila.infn.it
>>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #56643] error in fho file txt of PointStat
From: Paul Oldenburg
Time: Tue Jun 05 08:33:21 2012
Sabrina,
I'm afraid that there is a confusing explanation for the behavior you
reported. First of all, the statistic that you
are referring to as hit rate is calculated as PODY in the CTS output
line. Please refer to that. Regarding the FHO
H_RATE statistic, it was intended to be consistent with a legacy
verification system (the NCEP VSDB system). We realize
that the name H_RATE is confusing, but this is inherited behavior. I
removed the patch with the fixed version of FHO
H_RATE at the direction of my supervisors, so that is why the behavior
of your instance of MET did not change. Please
let me know if you have any questions.
Thanks,
Paul
On 06/05/2012 02:51 AM, Sabrina Gentile via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=56643>
>
> Dear Paul,
> I downloaded and reinstalled the patche
METv3.1_patches_20120501.tar.gz from
>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php
but
> I found again the previous problem in the FHO H_RATE statistic.
> Thanks, bye
>
> Sabrina
>
>
> ----- Original Message -----
> From: "Paul Oldenburg via RT"<met_help at ucar.edu>
> To:<sabrina.gentile at aquila.infn.it>
> Sent: Tuesday, May 29, 2012 8:19 PM
> Subject: Re: [rt.rap.ucar.edu #56643] error in fho file txt of
PointStat
>
>
>> Sabrina,
>>
>> A new patch for METv3.1 has been deployed to the MET website. You
can
>> download the patch here:
>>
http://www.dtcenter.org/met/users/support/known_issues/METv3.1/index.php.
>> The patched code should correctly calculate
>> the FHO H_RATE statistic. Thanks again for bringing this matter to
our
>> attention. If you have any questions or notice
>> any other problems, please let us know.
>>
>> Thanks,
>>
>> Paul
>>
>>
>> On 05/29/2012 11:12 AM, Paul Oldenburg wrote:
>>> Sabrina,
>>>
>>> Thank you for reporting this problem. It appears that you are
correct
>>> that the H_RATE calculation in the FHO line is
>>> incorrect. There is an equivalent calculation called PODY in the
CTC
>>> line, which appears to be calculated correctly.
>>> Please refer to this quantity until we are able to deploy a bug
fix for
>>> the error in the FHO H_RATE calculation. I will
>>> let you know when a bug fix is available for this problem.
>>>
>>> Thanks,
>>>
>>> Paul
>>>
>>>
>>> On 05/28/2012 07:52 AM, Sabrina Gentile via RT wrote:
>>>>
>>>> Mon May 28 07:52:20 2012: Request 56643 was acted upon.
>>>> Transaction: Ticket created by sabrina.gentile at aquila.infn.it
>>>> Queue: met_help
>>>> Subject: error in fho file txt of PointStat
>>>> Owner: Nobody
>>>> Requestors: sabrina.gentile at aquila.infn.it
>>>> Status: new
>>>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=56643>
>>>>
>>>>
>>>> Dear all,
>>>> I want to report an error in the FHO file output of PointStat
Tool in
>>>> MET3.1. The FHO txt file has in the 23 column
>>>> the F_RATE, in the 24 the H_RATE and in the 25 the O_RATE. I
suppose
>>>> that the calculation of the H_RATE is not
>>>> correct, in fact, if I control the n values in the contingency
table
>>>> from the CTC text file, I find that H_RATE in the
>>>> FHO text file is equal to n11/T (T=n11+n10+n01+n00) instead the
correct
>>>> H_RATE should be equal to n11/(n11+n01). In
>>>> the output file, in fact, I find H_RATE values very low due to
this
>>>> error.
>>>> I hope my suggestion will be helpful.
>>>> Best regards
>>>>
>>>>
>>>>
>>>>
>>>> Sabrina
>>>>
>>>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>> Sabrina Gentile, Ph.D
>>>> CETEMPS, University of L'Aquila
>>>> Department of Physics - Atmospheric Sciences
>>>> Via Vetoio, 67010 Coppito L'Aquila
>>>> Italy
>>>> E-mail: sabrina.gentile at aquila.infn.it
>>>
>>
>>
>
------------------------------------------------
More information about the Met_help
mailing list