# Position of an element in a vector

1,479 views (last 30 days)
F.O on 18 Oct 2017
Commented: Star Strider on 18 Oct 2017
Hi,I want to find the position not value of an element of a vector that meet equal a number, I have this vector z=[0:-0.01:-16] and want to find the position when z=-2 .

Cam Salzberger on 18 Oct 2017
Hello Farhan,
You can use find to get the index when an expression is true, and "==" to get that logical array. Something like this should work:
idx = find(z == -2);
Note, however, that this will only do an absolute comparison. If you wanted to find with some kind of tolerance, you could do:
idx = find(abs(z+2) < 0.001);
Also note that if you are then going to use this index for indexing into the array, it's usually more efficient to just use the logical array to index:
z(z == -2) = NaN; % Or something
Hope this helps!
-Cam

Show 1 older comment
Stephen Cobeldick on 18 Oct 2017
@Farhan Omar: here are the 201st and 21st elements:
>> z(21)
ans = -0.20000
>> z(201)
ans = -2
Note that Star Strider's answer using ismembertol will be much more robust than this answer.
F.O on 18 Oct 2017
I am sorry I had a mistake in the step which is 0.1 not 0.01 thanks for the answer and it make sense
Star Strider on 18 Oct 2017
@Stephen — Thank you!

Star Strider on 18 Oct 2017
Edited: Star Strider on 18 Oct 2017
The ismember (or perhaps preferably ismembertol) function is your friend here.
EDIT The ‘-2’ index will be 21 if the step in ‘z’ is ‘-0.1’ rather than ‘-0.01’:
z=[0:-0.1:-16];
[~,idx] = ismembertol(-2, z, 1E-8)
idx =
21

F.O on 18 Oct 2017
Thanks it works but it is difficult to understand how it works
Star Strider on 18 Oct 2017
The ismembertol function checks to see whether the first argument (here -2) is a member of the set in the second argument (here ‘z’), and returns the index/indices in the second argument that match. Since floating-point calculations (including those involved in the colon operator calculations) can produce values that are not exactly -2 (in this instance), the third ‘tolerance’ argument gives a range of values around -2 that would meet the criterion. The find and ismember functions match the condition exactly, while ismembertol matches within a tolerance.
See: Why is 0.3 - 0.2 - 0.1 (or similar) not equal to zero? (link) for a full explanation of the reason ismembertol may be the preferred function, if there is a reason to suspect that no element of ‘z’ is exactly equal to -2.

KL on 18 Oct 2017
Edited: KL on 18 Oct 2017
[row,col,v] = find(z==-2)

Jan on 18 Oct 2017
Remember the old FAQ:
Why does
any((0:0.1:1) == 0.3)