In ecdf (empirical cumulative distribution function), how can I find the indices of corresponding data point used for any chosen cumulative probability?

Hi everyone, I'm trying to find the indices of data points used in the ecdf function. Apparently, the original data used gets sorted, and I'm trying to get the original unsorted data points and match them to the probabilities found. Suppose I have [f1,bi] = ecdf(minbi). Once ecdf is calculated, data in bi gets sorted in ascending order, and so are the probabilities obtained in f1. How can I can 'unsort' vector bi and match the obtained probabilities to the corresponding original data points bi? Thanks in advance.

Answer 1

[the cyclist]

I think you could do what you want with

[tf loc] = ismember(minbi,bi);

loc will be the indices you want.

[Matlabio]

Thank you so much,it's working.