[ncl-talk] What is the method used in "dim_spi_n" function to estimate three gamma parameters of the Pearson type III distribution

[Mary Haley]

Kritanai,

Yes, by default, this function will calculate the SPI by applying a
2-parameter gamma distribution fit, regardless of whether you have NCL
Version 6.2.1 or 6.3.0.

To get the Pearson type III distribution, you need to have NCL V6.3.0 or
later, and you need to set:

  opt = True
  opt at spi_type = 3

You can see full information on the dim_spi_n page:

http://www.ncl.ucar.edu/Document/Functions/Built-in/dim_spi_n.shtml

--Mary


On Sun, Apr 24, 2016 at 4:44 AM, Kritanai Torsri <kritanai at haii.or.th>
wrote:

> Dear All,
>
> I used the "dim_spi_n" function with the Pearson type III to compute SPI.
>
> Am I correct that, by default, the NCL function still used the Maximum
> likelihood to estimate
>
> the location, scale, and shape of the gamma distribution as in
> two-parameter gamma distribution or it uses another method?
>
> If any one know, please give me suggestion and thanks in advance.
>
> BRGs,
>
> Kritanai Torsri
>
> Assistant Researcher
>
> Hydro and Agro Infomatics Institute, Bangkok, Thailand.
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20160425/7532bf46/attachment.html