[gradsusr] Contingency Table
Eric Altshuler
ela at cola.iges.org
Wed Apr 30 14:38:41 EDT 2014
Dear Amulya,
The grads scripting language does not support directly accessing the values of data variables (either defined or from a dataset). In your 'if' statements, the variables 'a' and 'b' refer to script variables, not data variables. You need to first obtain the values of your defined variables a and b, then assign these values to script variables. Just before each 'if' statement in your script, insert the following lines:
'display a'
a=subwrd(result,4)
'display b'
b=subwrd(result,4)
It may be a bit confusing, but the script variables 'a' and 'b' (i.e. those assigned values using subwrd function) are NOT the same as the data variables a and b (i.e. those in the 'display' commands).
Best regards,
Eric L. Altshuler
Research Scientist
Center for Ocean-Land-Atmosphere Studies
253 Research Hall, Mail Stop 6C5
George Mason University
4400 University Drive
Fairfax, VA 22030 USA
E-mail: ela at cola.iges.org
Phone: (703) 993-5725
Fax: (703) 993-5770
----- Original Message -----
From: "Amulya Chevuturi" <amulya.chevuturi at gmail.com>
To: "GrADS Users Forum" <gradsusr at gradsusr.org>
Sent: Wednesday, April 30, 2014 12:18:33 AM
Subject: [gradsusr] Contingency Table
Dear Grads Users
I am trying to build contingency table for model verification using grads. Currently my model and observation are both at exactly same resolution. So I have written the following script. But it is not giving me the count of Yes/Yes, No/No etc correctly. How am I supposed to build contingency table using grads if someone can please help.
Following is the script.
'reinit'
'open model.ctl'
'set t 3'
'set y 57 208'
'set x 72 228'
'define a = rain'
'close 1'
'open observation.ctl'
'set t 3'
'set y 57 208'
'set x 72 228'
'define b = prec'
n1=0
n2=0
n3=0
n4=0
j=72
while (j<=228)
'set y 'j
i=57
while (i<=208)
'set x 'i
if(a>0&b>0)
n1=n1+1
else
if(a<=0&b<=0)
n2=n2+1
else
if(a<=0&b>0)
n3=n3+1
else
if(a>0&b<=0)
n4=n4+1
endif
endif
endif
endif
i=i+1
endwhile
j=j+1
endwhile
say n1
say n2
say n3
say n4
'close 1'
Or is anyother method available for making contingency tables for forecast verification with grads?
Please let me know.
Regards
Amulya Chevuturi
_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org
http://gradsusr.org/mailman/listinfo/gradsusr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20140430/4d67bed3/attachment.html
More information about the gradsusr
mailing list