-
Notifications
You must be signed in to change notification settings - Fork 7
/
missmean.m
35 lines (30 loc) · 862 Bytes
/
missmean.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
function mm=missmean(X)
%MISSMEAN mean of a matrix X with NaN's
%
%[mm]=missmean(X)
%
%This function calculates the mean of a matrix X.
%X may hold missing elements denoted by NaN's which
%are ignored (weighted to zero).
%
%Check that for no column of X, all values are missing
% Copyright (C) 1995-2006 Rasmus Bro & Claus Andersson
% Copenhagen University, DK-1958 Frederiksberg, Denmark, rb@life.ku.dk
%Insert zeros for missing, correct afterwards
missidx = isnan(X);
i = find(missidx);
X(i) = 0;
%Find the number of real(non-missing objects)
if min(size(X))==1,
n_real=length(X)-sum(missidx);
else
n_real=size(X,1)-sum(missidx);
end
i=find(n_real==0);
if isempty(i) %All values are real and can be corrected
mm=sum(X)./n_real;
else %There are columns with all missing, insert missing
n_real(i)=1;
mm=sum(X)./n_real;
mm(i)=i + NaN;
end