Re: [ferret_users] calculating Temporal RMSE

[Arijeet Dutta <arijeet.uoh@xxxxxxxxx>]

Hi, ferret scripts are available for that, check this
../../faq/correlations-and-variances.html

Reg
Arijeet

On 10/5/16, afwande juliet <afwandej965@xxxxxxxxx> wrote:
> Thanks I will try this, does the same trick applies to correlation also
> since I would also want to calculate correlation and present results in
> table form
>
> On Oct 5, 2016 8:04 AM, "Arijeet Dutta" <arijeet.uoh@xxxxxxxxx> wrote:
>
 Hi, I guess I have confused you!

 let's try this, you said you have data for 1997, 12 months for 1997,
 i.e, the time axis shows you 12 points ( l 1:12)

 for MAM

 yes? let p = (obsrved-model)
 yes? let q = p^2
 yes? let r = q[l=3:5@sum] !for MAM
 yes? let s = r/3
 yes? let m = s^0.5
 yes? plot m

 For OND

 yes? let p = (obsrved-model)
 yes? let q = p^2
 yes? let r = q[l=10:12@sum] !for OND
 yes? let s = r/3
 yes? let m = s^0.5
 yes? plot m


 On 10/5/16, afwande juliet <afwandej965@xxxxxxxxx> wrote:
 > Thanks will try only that my data is for only 2 seasons MAM and OND
 1997, I
 > don't know whether it will accept now that you are talking about
 > climatology
 >
 > On Oct 5, 2016 7:54 AM, "Arijeet Dutta" <arijeet.uoh@xxxxxxxxx> wrote:
 >
  Hi,
  you can list the output
  e.g.
  yes?list/file = name.dat m ! 'm' is the desired rmse following last
  mail
 
  On 10/5/16, afwande juliet <afwandej965@xxxxxxxxx> wrote:
  > Thanks But my aim is to get the values in kind of table or something
  >
  > On Oct 5, 2016 7:00 AM, "Arijeet Dutta" <arijeet.uoh@xxxxxxxxx>
 wrote:
  >
   !say you want it for JJAS and you have the climatology
   yes? let p = (obsrved-model)
   yes? let q = p^2
   yes? let r = q[l=6:9@sum] !for JJAS
   yes? let s = r/4
   yes? let m = s^0.5
   yes? plot m
  
   Regards
   Arijeet
  
   On 10/4/16, afwande juliet <afwandej965@xxxxxxxxx> wrote:
   > Dear All
   > I want to calculate temporal root mean square error between the
   > Observed
   > and Model output for different seasons. I have tried to search
   > here
  but
   > I
   > dont get clear way to do. Kindly if someone has an idea how to .
   >
   >
  
  >
 
 >

>