[Met_help] Re: funny HK example

[John Halley Gotway]

Dave,

Please replace the file METv1.1/lib/vx_contable/vx_contable_stats.cc with the attached version and recompile MET.

Then when you rerun, the correct value of HK should be computed.

Thanks for finding this issue!

(MET-Help folks... fyi, Dave found a bug in the computation of the Hanssen and Kuipers discriminant for a case he was running.  The formula in the code is correct but we were having integer overflow
problems.)

John

David Ahijevych wrote:
> Here's the observation and forecast that produce a bad HK.
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vx_contable_stats.cc
Type: text/x-c++src
Size: 9420 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/met_help/attachments/20081125/d4e0e7ef/vx_contable_stats.bin