Re: standard deviation and related coefficient

To read in ASCII files that potentially has no values in some columns, you would normally use asciiread to read in the data as strings, and then use str_get_cols to parse out the columns you want.

Your file has some tabs in it, though, which makes the columns uneven. Your best bet might be to write a Fortran program to read this file in, and then wrap it with WRAPIT so you can call it from NCL:

http://www.ncl.ucar.edu/Document/Tools/WRAPIT.shtml

For all the other questions, please read the documentation. There are functions for averaging, standard deviation, etc:

http://www.ncl.ucar.edu/Document/Functions/math.shtml

--Mary

On Feb 3, 2012, at 1:07 AM, nlcheng wrote:

> Supose that I have three colum data such as below:
>
> time U_M U
> 1 7.216151 9.8
> 2 　 10.5
> 3 9.295473 8.4
> 4 　 8.9
> 5 9.411055 8.5
> 6 9.293616 　
> 7 9.138997 8.6
> 8 8.930552 7.4
> 9 8.597112 5.4
> 10 8.098866 6.7
> 11 6.489672 　
> 12 6.111356 5.2
> 13 　 3.1
> 14 5.382022 3.9
> 15 4.988408 3.3
> you can see that some data is missing.
> how can i write the command to caculate the average value of U_M ?
> and the standard deviation <截图1.png>
> and the related coefficient of U_M ?
> it is the same to U.
>
> further on how can i write the command to caculate the t when a is 0.05?
> how can i caculate the portfolio concentration such as <截图3.png>2% is 76%?
> thank you very much!
>
>
>
>
>
>
>
>
> <截图2.png>_______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Mon Feb 6 16:01:13 2012