Hi! I am trying to find a way to calculate the 90% percentile of a matrix (1,n) . I provide an example below, and the value I would need to find is 0.292. (Note the matrix will changes in length). How could this be done?
[0.289, 0.254, 0.287, 0.292, 0.289, 0.267, 0.289, 0.304, 0.284, 0.282]

 Réponse acceptée

Star Strider
Star Strider le 17 Août 2020

3 votes

Here are two possibilities, the second of which gives you the exact value in your Question:
v = [0.289, 0.254, 0.287, 0.292, 0.289, 0.267, 0.289, 0.304, 0.284, 0.282];
Out1 = prctile(v, 90)
Out2 = interp1(linspace(1/numel(v),1,numel(v)), sort(v), 0.9)
producing:
Out1 =
0.2980
Out2 =
0.2920
.

6 commentaires

Jonathan Moorman
Jonathan Moorman le 17 Août 2020
Out2 was exactly what I was looking for. Thanks!
Star Strider
Star Strider le 17 Août 2020
As always, my pleasure!
Jonathan Moorman
Jonathan Moorman le 1 Sep 2020
Hi Star,
I've just now run into a problem with the line of code you sent. Until now it was working perfectly, but its currently giving a 90th percentile value that is not in the matrix. I gave a screenshot of the problem. Is there something incorrect with my code or is this something you have seen before? Thanks
I do not have the vector to work with. (I can barely read the images of it.)
The 90th percentile may not actually be an element of the vector, simply what that element would be if it were there, since that by default uses linear interpolation.
See if this version does what you want, with the interpolation method now 'nearest':
Out2 = interp1(linspace(1/numel(v),1,numel(v)), sort(v), 0.9, 'nearest')
An anonymous function version of it would be:
prctlv = @(v) interp1(linspace(1/numel(v),1,numel(v)), sort(v), 0.9, 'nearest');
That will make it easier to work with. Call it as:
a90Per = prctlv(a90Export);
I tested that with the original ‘v’ and it again gave the desired result.
Jonathan Moorman
Jonathan Moorman le 3 Sep 2020
Working great again. Thanks!
Star Strider
Star Strider le 3 Sep 2020
As always, my pleasure!

Connectez-vous pour commenter.

Plus de réponses (1)

jonas
jonas le 17 Août 2020

1 vote

There is probably a one-liner for this, but I guess you could use
A = [0.289, 0.254, 0.287, 0.292, 0.289, 0.267, 0.289, 0.304, 0.284, 0.282];
B = sort(A);
id = round(numel(A).*0.9)
B(id)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by