Re: [ferret_users] cumulative distribution curve

[Ryo Furue <furue@xxxxxxxxxx>]

On Mon, Dec 2, 2019 at 5:33 PM Ryo Furue <furue@xxxxxxxxxx> wrote:

Gopika,

On Mon, Dec 2, 2019 at 3:22 PM gopika suresh <gopikasurvay@xxxxxxxxx> wrote:

DEFINE AXIS/X occurences = hval[x=1:33]

I don't know what hval is.  Also, has it been already sorted?  See

     https://www.pmel.noaa.gov/maillists/tmap/ferret_users/fu_2001/msg00631.html

 

LET pdf = I[gx=occurences]

Please you the lower-case "i": I thought it was an ell (L)!

 

DEFINE AXIS/X=1900:2100:1 my_intervals

let frequency=pdf[gx=my_intervals@sum]

I guess this gives you a PDF (histogram) of hval, but the value at x=2019, for example, is the number of occurrences between 2018.5 and 2019.5.  If it's okay, then I think this is what you want.

Oh, and I've found your data in data.nc is defined on the Z axis.  In the example above, you use X.

 

I don't find other problems.  I hope people more familiar with this kind of operations will be able to help.

Cheers,

Ryo