[ncl-talk] Remove trend from a data having non-linear trend

Dennis Shea shea at ucar.edu
Tue Jun 27 12:19:23 MDT 2017


A very simple exponential fit function is attached. A test driver is also
provided. Both were recently provided in an ncl-talk response to a query.
You will have to develop your own exponential fit if you want something
more sophisticated.

%> ncl exponential_curve_fit.ncl

Good Luck

On Tue, Jun 27, 2017 at 9:59 AM, S Br <sbr.climate at gmail.com> wrote:

> Hi,
> Thanks for your reply. I will try to use these function.
> I understood you have suggested the polynomial fit function. Is there any
> 'exponential fit' function available to remove the non-linear trend.
>
> SB
>
> On Tue, Jun 27, 2017 at 3:09 PM, Dennis Shea <shea at ucar.edu> wrote:
>
>> There is no explicit function to remove some unknown non-linear response.
>>
>> You must ascertain (eg: eye-ball) the 'shape' of the non-linear response
>> and use a function: eg
>>    http://www.ncl.ucar.edu/Document/Functions/Built-in/lspoly_n.shtml
>> to generate some baseline. You can remove the non-linear response via
>> (data-lspoly)
>> Note: you should plot the output from lspoly_n to make sure it is
>> appropriate.
>>
>> An example of using lspoly is here:
>> http://www.ncl.ucar.edu/Applications/regress.shtml#ex6
>>
>>
>>
>> On Mon, Jun 26, 2017 at 10:20 AM, SBR climate <sbr.climate at gmail.com>
>> wrote:
>>
>>> Hi NCL users,
>>> I have generated an area averaged time series where the data is having
>>> non-linear trend.Is there any function available to to remove the trend
>>> from the time-series having non-linear trend.
>>>
>>> Thanks
>>> SB
>>>
>>> _______________________________________________
>>> 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/20170627/0c38f0d3/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: exponential_curve_fit.ncl
Type: application/octet-stream
Size: 4610 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20170627/0c38f0d3/attachment.obj 


More information about the ncl-talk mailing list