# How can I make multiple max in elegant way?

3 views (last 30 days)
Eli Borodach on 10 Mar 2016
Commented: Walter Roberson on 10 Mar 2016
Hello all, let's assume I have 8 vectors x1,x2,x3,...,x8 I can use: max(max(max(x1,x2),x3),x4) and so on... But how can I do it in a more elegant way?

Ced on 10 Mar 2016
Edited: Ced on 10 Mar 2016
you could do
max(max([ x1 x2 x3 x4 x5 .... ]))
Since the first max is evaluated along the columns, I think it should be fast enough. Alternatively, you could of course do X = [ x1 x2 x3 ... ] and then max(X(:)). That's shorter, but depending on the length of your vectors, I would go with the first option. A common alternative is to use sort, e.g.
X_sorted = sort([ x1 x2 ... ]);
xmax = max(X_sorted(end,:))
but since for max, you don't need the matrix to be fully sorted, I would guess that max is actually faster.
##### 2 CommentsShowHide 1 older comment
Walter Roberson on 10 Mar 2016
Are you looking for the overall maximum among the 8 vectors? Or are you looking for the maximum for each entry in the vector -- e.g.,
[max([x1(1), x2(1), x3(1), x4(1), x5(1), x6(1), x7(1), x8(1)]),
max([x1(2), x2(2), x3(2), x4(2), x5(2), x6(2), x7(2), x8(2)]),
...
max([x1(end), x2(end), x3(end), x4(end), x5(end), x6(end), x7(end), x8(end)]) ];
Your max(max(max(...max(x1,x2), x3), x4 ... code is taking maximum for each entry, which is what I coded for. Some of the solutions Ced posted are for the overall maximum.

Walter Roberson on 10 Mar 2016
If they have the same orientation but the orientation is not known ahead of time,
max( cat(3, x1, x2, x3, x4, x5, x6, x7, x8), [], 3)
The result would have the same orientation.
Or you could use
max( [x1(:), x2(:), x3(:), x4(:), x5(:), x6(:), x7(:), x8(:)], [], 2)
the result would be a column vector.
If you know that they are column vectors already you can leave out the (:)
Eli Borodach on 10 Mar 2016
Thank you very much!