Creating an indicator variable based on missing variable values

[ajb]

Hi,

I want to create a missing indicator variable and I am at a loss as to how to go about doing this.

I have 100 obs(under the variable PatID) and 15 other variables(v1-v15)

What I want to be able to do is:

        if an observation has missing values for at least any 3  of the 15 variables then newvar=1

else  if an observation has missing values for less than any 3 of the 15 variables then newvar=0

Any suggestions?

Thanks!!

[Jagadishkatam · Accepted solution]

you can use the nmiss function to achieve your output.

something like below

if nmiss(of v1-v15)>=3 then newvar=1;

else if nmiss(of v1-v15)<3 then newvar=0;

Thanks,

Jag

[ajb]

Thank you so much!