Re: select avg (ignore null)

[davidcrossey]

Hi powerpeanuts,

As per this note avg does ignore nulls for a single vector, so you could apply the average to each pair of values from each column as follows:

q)t:([]c1:`a`b`c;c2:1 0N 2;c3:5 3 5) 
q)update c4:avg each flip (c2;c3) from t 
c1 c2 c3 c4 
------------ 
a  1  5  3 
b     3  3 
c  2  5  3.5

Hope this helps