[ncl-talk] question concerning reg_multlin_stats

[Daniel Barandiaran]

Sorry for the bump, but I've been thinking about this a little more, and
looked at some more of contributed.ncl. I think that I have a simple
solution, which is just to use the ind function to pre-treat my data to
ignore missing values before feeding it to the regression function.

If this is a poor solution, or there is a better one than feel free to
chime in.

On Tue, Dec 1, 2015 at 8:36 AM, Daniel Barandiaran <dbarandiaran at gmail.com>
wrote:

> Hi everyone,
>
> In the documentation, it states that missing values are allowed, but not
> recommended for independent variables. However, when I give the function a
> time series for the dependent variable that has missing values, it says
> this is not allowed and dumps out. I peeked into contributed.ncl, found the
> function and commented out the missing value check for the dependent
> variable. The function seems to work properly. Is this ok?
>
> I am using this function to reconstruct snow course SWE measurements with
> tree-rings. Some snow courses have a missing value or two in their record,
> but still have sufficient number of observations to result in reasonable
> regression models, and so I would still like to be able to use them.
>
> --
> Danny Barandiaran
> Ph.D. Candidate
> Department of Plants, Soils, and Climate
> Utah State University
>
>
>


-- 
Danny Barandiaran
Ph.D. Candidate
Department of Plants, Soils, and Climate
Utah State University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20151201/dc03eb9e/attachment.html