# Re: [ferret_users] Co relation coefficient.

Hi Nicolas,

Thanks for your suggestion, please go through the mails which we have shared i do understand that i have apply all these concepts and have to work on my own i do get that but the thing is, this is not the first time i have encountered this problem and till date i have not got any solution and I am not complaining about any thing over here the suggestions which i got i tried to implement and when someone has asked me something so i think clearing that is very important as Ansley,C Kalyani and samir gave me an indication that whether temperature is present in the data set so i just cleared that, Temperature is actually present in the data set, and i replied as it is not working so that i can any other resolution for this problem.

Please don't mind me saying that, the reply which you gave shows me anger so please try to avoid it.

thanks alot for your help i know you guys are doing a great job by helping us to work with ferret.

Thanks regards,
satyam.

On Thu, May 28, 2015 at 12:54 PM, Nicolas Freychet wrote:
Please, do not just copy the tips and complain it doesn't work. The solution the people give to you are juste example, you need to adapt to your case (and for that, you need to think about what you are doing).

You just said "i don't know why this is taking up the Temperature from the second data set". Well, it is because you specify Temperature should be in data 2 (let q = temperature[y=35,d=2]  ) So please, before sending a question, try to think a little by yourself why it's not working (I know sometimes it's difficult, but it actually help a lot to think on the problem before asking for help, you will learn a lot more this way).

Best,
Nicolas

2015-05-28 15:02 GMT+08:00 Satyam Srivastava :
Hi all thanks for your kind reply, i have been using the same data set for all kind of dummy computation but in this case i don't know why this is taking up the Temperature from the second data set the files are as follows ::

yes? set mode metafile
yes? set data pottm2008.nc
yes? set data pottmp.2008.nc
yes? let p = pottmp[y=35,d=1]
yes? let q = temperature[y=35,d=2]
yes? go variance
... Variance and Covariance: Instructions:
Use the LET/QUIET command to define the variable(s) P (and Q) as
your variable(s) of interest (e.g. yes? LET/QUIET P = u[x=180,y=0])
The variance of P will be variable P_VAR  (Q --> Q_VAR)
The covariance will be COVAR The correlation will be CORREL.
Type GO VAR_N to obtain n/n+1 statistical correction factor
...
yes? go var_n
CANCEL MODE VERIFY
**ERROR: variable unknown or not in data set: TEMPERATURE[Y=35,D=2]
IF `nm1 eq 0` then exit/script
Command file, command group, or REPEAT execution aborted
yes? use pottmp.2008.nc
yes? sh d
currently SET data sets:
1> ./pottm2008.nc
name     title                             I         J         K         L         M         N
TEMPERATURE
potential temperature (degc)     1:2001    1:1093    1:10      1:32      ...       ...

2> ./pottmp.2008.nc  (default)
name     title                             I         J         K         L         M         N
DATE                                      ...       ...       ...       1:12      ...       ...
TIMEPLOT                                  ...       ...       ...       1:12      ...       ...
POTTMP   Potential temperature            1:360     1:418     1:40      1:12      ...       ...

yes? use pottm2008.nc
yes? sh d
currently SET data sets:
1> ./pottm2008.nc  (default)
name     title                             I         J         K         L         M         N
TEMPERATURE
potential temperature (degc)     1:2001    1:1093    1:10      1:32      ...       ...

the terms are present in the file but it is unable to be read,

please have a look to the data set and please suggest me any solution if it possible.

Regards,
Satyam srivastava.

On Wed, May 27, 2015 at 10:16 PM, Ansley Manke wrote:
Hi,
The error message is telling you that there is not a variable named temperature in dataset 2. Use SHOW DATA to list the variables in the two datasets and make sure your LET definitions match which datset each variable is in.

On 5/27/2015 3:25 AM, Satyam Srivastava wrote:
Thanks Chinnu and Nicolas for your kind reply and help but after trying that as well then also i am getting the same error.

yes? set mode metafile
yes? set data pottm2008.nc
yes? set data pottmp.2008.nc
yes? let p = pottmp[y=35,d=1]
yes? let q = temperature[y=35,d=2]
yes? go variance
... Variance and Covariance: Instructions:
Use the LET/QUIET command to define the variable(s) P (and Q) as
your variable(s) of interest (e.g. yes? LET/QUIET P = u[x=180,y=0])
The variance of P will be variable P_VAR  (Q --> Q_VAR)
The covariance will be COVAR The correlation will be CORREL.
Type GO VAR_N to obtain n/n+1 statistical correction factor
...
yes? go var_n
CANCEL MODE VERIFY
**ERROR: variable unknown or not in data set: TEMPERATURE[Y=35,D=2]
IF `nm1 eq 0` then exit/script
Command file, command group, or REPEAT execution aborted

Regards,
Satyam

On Wed, May 27, 2015 at 3:33 PM, Nicolas Freychet wrote:
You need to precise the data of each variable (because you use different data) :
let p = pottmp[y=35,d=1]
let q = temperature[y=35,d=2]
(or switch d=1 and d=2, and don't know which of your data contains the variables)

2015-05-27 17:23 GMT+08:00 Satyam Srivastava :

yes? set mode metafile
yes? set data pottm2008.nc
yes? set data pottmp.2008.nc
yes? let p = pottmp[y=35]
yes? let q = temperature[y=35]
yes? go variance
... Variance and Covariance: Instructions:
Use the LET/QUIET command to define the variable(s) P (and Q) as
your variable(s) of interest (e.g. yes? LET/QUIET P = u[x=180,y=0])
The variance of P will be variable P_VAR  (Q --> Q_VAR)
The covariance will be COVAR The correlation will be CORREL.
Type GO VAR_N to obtain n/n+1 statistical correction factor
...
yes? go var_n
CANCEL MODE VERIFY
**ERROR: variable unknown or not in data set: TEMPERATURE[Y=35]
IF `nm1 eq 0` then exit/script
Command file, command group, or REPEAT execution aborted

On Wed, May 27, 2015 at 12:01 PM, samir pokhrel wrote:
Dear Satyam,

Make Sure that your p and q dimensions matches, it seems from your code that longitude of p vary from 0:30N and that of q from 0 to 1N. Please correct it and verify

Regards
Samir

On Wed, May 27, 2015 at 11:39 AM, Satyam Srivastava wrote:
Hi Folks,

I am trying to calculate the co relation coefficient for two different data set I have already gone through the ferret FAQ, from there i got the idea to compute for a single data set can any one help me for two different data set on different grids and they are time series data.

as i am trying the same concept but i am getting the errors as follows:

yes? set mode metafile
yes? use pottm2008.nc
yes? use pottmp.2008.nc
yes? let p=temperature[x=0:30,y=0:35]
yes? let q=pottmp[x=0:1,y=0:35]
yes? go var_n
CANCEL MODE VERIFY
**ERROR: variable unknown or not in data set: MASK_PQ
IF `nm1 eq 0` then exit/script
Command file, command group, or REPEAT execution aborted
yes?

--
Regards,

Satyam Srivastava,
J.R.F,
K. Banerjee Centre of Ocean Sciences,
+918756123444

--
Samir Pokhrel
Climate and Global Modelling Division
Indian Institute of Tropical Meteorology
NCL Post, Pashan
Pune-411008
INDIA

--
Regards,

Satyam Srivastava,
J.R.F,
K. Banerjee Centre of Ocean Sciences,
+918756123444

--
----------------------------------------------------
Nicolas Freychet
Postdoc Reseach Fellow at RCEC
----------------------------------------------------

--
Regards,

Satyam Srivastava,
J.R.F,
K. Banerjee Centre of Ocean Sciences,
+918756123444

--
Regards,

Satyam Srivastava,
J.R.F,
K. Banerjee Centre of Ocean Sciences,
+918756123444

--
----------------------------------------------------
Nicolas Freychet
Postdoc Reseach Fellow at RCEC
----------------------------------------------------

--
Regards,

Satyam Srivastava,
J.R.F,
K. Banerjee Centre of Ocean Sciences,