Re: missings in logistic regression

[Alan Mead]

How are you defining missing values? I notice that the handling of
missing cases depends on their status as system or user-defined. Maybe
you've found a bug.

https://www.gnu.org/software/pspp/manual/html_node/LOGISTIC-REGRESSION.html

-Alan

On 4/21/2017 10:16 AM, ftr public wrote:
> Hi,
>
> I calculated a logistic regression based on poll data that show a
> large percentage of missing data.
>
> The initially summary shows - the missing sub-command set on default -
>
> included in the analysis    8,196
> missing cases    7,253
> Total    15,449
>
> To see whether the large part of missings has any effect on the result
> I excluded missings by counting all missing values in the list of
> dependent and independent variables and then filtered for no missings.
> In addition I set the missing sub-command on exclude.
>
> count ka =   list of independent variables  (missing values).
> temp.
> sel if ka = 0.
> LOGISTIC REGRESSION dependent WITH   independent
>     /CATEGORICAL =     ...
>     /CRITERIA = CUT(0.5) ITERATE(20)
>     /NOORIGIN
>     /miss exclude.
>
> To my surprise, the number of missings did not drop to 0 .
>
> The summary now shows
>
> included in the analysis    8,196
> missing cases    4,200
> Total    12,396
>
> How can I achieve to get the missings excluded so that included total ?
>
> - ftr
>
>
> _______________________________________________
> Pspp-users mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/pspp-users

-- 

Alan D. Mead, Ph.D.
President, Talent Algorithms Inc.

science + technology = better workers

http://www.alanmead.org

I've... seen things you people wouldn't believe...
functions on fire in a copy of Orion.
I watched C-Sharp glitter in the dark near a programmable gate.
All those moments will be lost in time, like Ruby... on... Rails... Time for Pi.

          --"The Register" user Alister, applying the famous 
            "Blade Runner" speech to software development