[Met_help] [rt.rap.ucar.edu #38135] History for problem with MAE stat_analysis
RAL HelpDesk {for Paul Oldenburg}
met_help at ucar.edu
Fri Jun 4 10:44:15 MDT 2010
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi!
I have some problems with getting a value for MAE in stat_analysis. I've
made for different forecast steps of a WRF run Point_stat output and now
I'm trying to aggregate this into standard statistics like bias, ME,
RMSE, MAE etc. For all of those statistics I get a valid number, but not
for MAE.
In the link below you find the output stat-files from point_stat and
also from stat_analysis (with the NaN for MAE and values for the other
statistics). Can you help me to find the solution for this?
Cheers,
Marlous Jonker
The job I'm trying to run is:
/cm/shared/apps/met/2.0/METv2.0/bin/stat_analysis -lookin
${maindir}/out/point_stat/wrf_9km_eu_20100526_06\
-job aggregate_stat -line_type sl1l2 -out_line_type CNT -fcst_var TMP
\
-out ${maindir}/out/stat_analysis/20100526/wrf_9km_eu_20100526_06
_tmp.2.txt -v 2
The link to the data is:
https://www.wetransfer.com/dl.php?code=375aAdT4&hash=dda54e9ecfaaf30430b
266cd943d93e9e1c321cbaa474fe9d8f5a46c0b01aa145d21fe28943cc
-------------------------------------------------------------------
Marlous Jonker
Meteorological Researcher
Meteo Consult BV, Wageningen, the Netherlands
Tel: +31 (0)317 399872
Email: M.Jonker at weer.nl
www: research.meteogroup.com
-------------------------------------------------------------------
--
This e-mail is from Meteo Consult B.V., a MeteoGroup company. For more information, see http://www.weer.nl/gebruiksvoorwaarden.
This e-mail may contain confidential information. Only the addressee is permitted to read, copy, distribute or otherwise use this e-mail or any attachments. If you have received it in error, please contact the sender immediately. Any opinion expressed in this e-mail is personal to the sender and may not reflect the opinion of MeteoGroup.
Any e-mail reply to this address may be subject to interception or monitoring for operational reasons or for lawful business practices.
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #38135] problem with MAE stat_analysis
From: Paul Oldenburg
Time: Fri May 28 12:43:47 2010
Marlous,
We see that you are trying to aggregate sl1l2 partial sum lines and
compute
continuous statistics from them. Unfortunately, this procedure only
works for a
subset of the continuous statistics due to mathematical
technicalities. The
only way to generate MAE would be to have the MET point_stat tool
output the
matched pair (MPR) line type. Then, use stat_analysis on those lines
to produce
MAE through time. If you have any questions, please let us know.
Paul
RAL HelpDesk {for m.jonker at weer.nl} wrote:
> Fri May 28 06:03:03 2010: Request 38135 was acted upon.
> Transaction: Ticket created by m.jonker at weer.nl
> Queue: met_help
> Subject: problem with MAE stat_analysis
> Owner: Nobody
> Requestors: m.jonker at weer.nl
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38135 >
>
>
> Hi!
>
>
>
> I have some problems with getting a value for MAE in stat_analysis.
I've
> made for different forecast steps of a WRF run Point_stat output and
now
> I'm trying to aggregate this into standard statistics like bias, ME,
> RMSE, MAE etc. For all of those statistics I get a valid number, but
not
> for MAE.
>
> In the link below you find the output stat-files from point_stat and
> also from stat_analysis (with the NaN for MAE and values for the
other
> statistics). Can you help me to find the solution for this?
>
>
>
> Cheers,
>
> Marlous Jonker
>
>
>
> The job I'm trying to run is:
>
>
>
> /cm/shared/apps/met/2.0/METv2.0/bin/stat_analysis -lookin
> ${maindir}/out/point_stat/wrf_9km_eu_20100526_06\
>
> -job aggregate_stat -line_type sl1l2 -out_line_type CNT -fcst_var
TMP
> \
>
> -out ${maindir}/out/stat_analysis/20100526/wrf_9km_eu_20100526_06
> _tmp.2.txt -v 2
>
>
>
> The link to the data is:
>
>
https://www.wetransfer.com/dl.php?code=375aAdT4&hash=dda54e9ecfaaf30430b
> 266cd943d93e9e1c321cbaa474fe9d8f5a46c0b01aa145d21fe28943cc
>
>
>
>
>
> -------------------------------------------------------------------
>
> Marlous Jonker
>
> Meteorological Researcher
>
> Meteo Consult BV, Wageningen, the Netherlands
>
> Tel: +31 (0)317 399872
>
> Email: M.Jonker at weer.nl
>
> www: research.meteogroup.com
>
> -------------------------------------------------------------------
>
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #38135] problem with MAE stat_analysis
From: m.jonker at weer.nl
Time: Mon May 31 00:43:44 2010
Hi!
Thanks for your help. We are now running point_stat to create MPR
output and then calculate the MAE from that.
What I don't understand however is how the bias is calculated and the
absolute error is not, even when it is calculated from the same data.
Can you please explain this?
Thanks again for your help.
Cheers,
Marlous
-----Oorspronkelijk bericht-----
Van: RAL HelpDesk {for Paul Oldenburg} [mailto:met_help at ucar.edu]
Verzonden: vrijdag 28 mei 2010 20:44
Aan: Marlous Jonker
Onderwerp: Re: [rt.rap.ucar.edu #38135] problem with MAE stat_analysis
Marlous,
We see that you are trying to aggregate sl1l2 partial sum lines and
compute
continuous statistics from them. Unfortunately, this procedure only
works for a
subset of the continuous statistics due to mathematical
technicalities. The
only way to generate MAE would be to have the MET point_stat tool
output the
matched pair (MPR) line type. Then, use stat_analysis on those lines
to produce
MAE through time. If you have any questions, please let us know.
Paul
RAL HelpDesk {for m.jonker at weer.nl} wrote:
> Fri May 28 06:03:03 2010: Request 38135 was acted upon.
> Transaction: Ticket created by m.jonker at weer.nl
> Queue: met_help
> Subject: problem with MAE stat_analysis
> Owner: Nobody
> Requestors: m.jonker at weer.nl
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38135 >
>
>
> Hi!
>
>
>
> I have some problems with getting a value for MAE in stat_analysis.
I've
> made for different forecast steps of a WRF run Point_stat output and
now
> I'm trying to aggregate this into standard statistics like bias, ME,
> RMSE, MAE etc. For all of those statistics I get a valid number, but
not
> for MAE.
>
> In the link below you find the output stat-files from point_stat and
> also from stat_analysis (with the NaN for MAE and values for the
other
> statistics). Can you help me to find the solution for this?
>
>
>
> Cheers,
>
> Marlous Jonker
>
>
>
> The job I'm trying to run is:
>
>
>
> /cm/shared/apps/met/2.0/METv2.0/bin/stat_analysis -lookin
> ${maindir}/out/point_stat/wrf_9km_eu_20100526_06\
>
> -job aggregate_stat -line_type sl1l2 -out_line_type CNT -fcst_var
TMP
> \
>
> -out ${maindir}/out/stat_analysis/20100526/wrf_9km_eu_20100526_06
> _tmp.2.txt -v 2
>
>
>
> The link to the data is:
>
>
https://www.wetransfer.com/dl.php?code=375aAdT4&hash=dda54e9ecfaaf30430b
> 266cd943d93e9e1c321cbaa474fe9d8f5a46c0b01aa145d21fe28943cc
>
>
>
>
>
> -------------------------------------------------------------------
>
> Marlous Jonker
>
> Meteorological Researcher
>
> Meteo Consult BV, Wageningen, the Netherlands
>
> Tel: +31 (0)317 399872
>
> Email: M.Jonker at weer.nl
>
> www: research.meteogroup.com
>
> -------------------------------------------------------------------
>
>
>
>
>
--
This e-mail is from Meteo Consult B.V., a MeteoGroup company. For more
information, see http://www.weer.nl/gebruiksvoorwaarden.
This e-mail may contain confidential information. Only the addressee
is permitted to read, copy, distribute or otherwise use this e-mail or
any attachments. If you have received it in error, please contact the
sender immediately. Any opinion expressed in this e-mail is personal
to the sender and may not reflect the opinion of MeteoGroup.
Any e-mail reply to this address may be subject to interception or
monitoring for operational reasons or for lawful business practices.
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #38135] problem with MAE stat_analysis
From: Paul Oldenburg
Time: Tue Jun 01 10:32:21 2010
Marlous,
I assume you are referring to multiplicative bias when you say bias?
If so,
multiplicative bias can be calculated by combining the information
stored in the
SL1L2 lines. There is more information on how to do this on page C-7
of the MET
User's Guide. The information required to calculate MAE is not
present in the
SL1L2 lines. Specifically, that calculation requires the sum of the
absolute
errors which are not present (see page C-8).
You can aggregate the MAE values in the CNT lines of the MET output by
simply
taking a weighted average using the TOTAL field as the weight. If you
have any
questions, please let us know.
Thanks,
Paul
RAL HelpDesk {for m.jonker at weer.nl} wrote:
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38135 >
>
> Hi!
>
> Thanks for your help. We are now running point_stat to create MPR
output and then calculate the MAE from that.
> What I don't understand however is how the bias is calculated and
the absolute error is not, even when it is calculated from the same
data. Can you please explain this?
>
> Thanks again for your help.
>
> Cheers,
> Marlous
>
> -----Oorspronkelijk bericht-----
> Van: RAL HelpDesk {for Paul Oldenburg} [mailto:met_help at ucar.edu]
> Verzonden: vrijdag 28 mei 2010 20:44
> Aan: Marlous Jonker
> Onderwerp: Re: [rt.rap.ucar.edu #38135] problem with MAE
stat_analysis
>
> Marlous,
>
> We see that you are trying to aggregate sl1l2 partial sum lines and
compute
> continuous statistics from them. Unfortunately, this procedure only
works for a
> subset of the continuous statistics due to mathematical
technicalities. The
> only way to generate MAE would be to have the MET point_stat tool
output the
> matched pair (MPR) line type. Then, use stat_analysis on those
lines to produce
> MAE through time. If you have any questions, please let us know.
>
> Paul
>
> RAL HelpDesk {for m.jonker at weer.nl} wrote:
>> Fri May 28 06:03:03 2010: Request 38135 was acted upon.
>> Transaction: Ticket created by m.jonker at weer.nl
>> Queue: met_help
>> Subject: problem with MAE stat_analysis
>> Owner: Nobody
>> Requestors: m.jonker at weer.nl
>> Status: new
>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=38135 >
>>
>>
>> Hi!
>>
>>
>>
>> I have some problems with getting a value for MAE in stat_analysis.
I've
>> made for different forecast steps of a WRF run Point_stat output
and now
>> I'm trying to aggregate this into standard statistics like bias,
ME,
>> RMSE, MAE etc. For all of those statistics I get a valid number,
but not
>> for MAE.
>>
>> In the link below you find the output stat-files from point_stat
and
>> also from stat_analysis (with the NaN for MAE and values for the
other
>> statistics). Can you help me to find the solution for this?
>>
>>
>>
>> Cheers,
>>
>> Marlous Jonker
>>
>>
>>
>> The job I'm trying to run is:
>>
>>
>>
>> /cm/shared/apps/met/2.0/METv2.0/bin/stat_analysis -lookin
>> ${maindir}/out/point_stat/wrf_9km_eu_20100526_06\
>>
>> -job aggregate_stat -line_type sl1l2 -out_line_type CNT -fcst_var
TMP
>> \
>>
>> -out ${maindir}/out/stat_analysis/20100526/wrf_9km_eu_20100526_06
>> _tmp.2.txt -v 2
>>
>>
>>
>> The link to the data is:
>>
>>
https://www.wetransfer.com/dl.php?code=375aAdT4&hash=dda54e9ecfaaf30430b
>> 266cd943d93e9e1c321cbaa474fe9d8f5a46c0b01aa145d21fe28943cc
>>
>>
>>
>>
>>
>> -------------------------------------------------------------------
>>
>> Marlous Jonker
>>
>> Meteorological Researcher
>>
>> Meteo Consult BV, Wageningen, the Netherlands
>>
>> Tel: +31 (0)317 399872
>>
>> Email: M.Jonker at weer.nl
>>
>> www: research.meteogroup.com
>>
>> -------------------------------------------------------------------
>>
>>
>>
>>
>>
>
>
>
>
------------------------------------------------
More information about the Met_help
mailing list