Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I asked a question exactly like this a while ago, so I do not know if it is appropriate to ask pretty much the same question with a single tweak.

For the record, my first question is In an N-dimensional space filled with points, systematically find the closest point to a specified point

Now I would like to use spearmans correlation rather than using euclidean distance. (http://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient)

I tried using my method that I described in my other question with the spearmans correlation. I generated a random list of 10 points and then an extra point to compare with. I did the analysis on this set and I got this data:

Spearmans Correlated Rank: Program rank

1:2

2:5

3:7

4:9

5:1

6:3

7:6

8:8

9:4

10:10

which shows that my method will not work at all. Is there a suggested way of going about this with spearmans correlation?

share|improve this question
add comment

1 Answer

up vote 0 down vote accepted

If I understand correctly, Spearman rank correlation depends solely on Euclidean distance between rank vectors. Therefore, you can simply apply your nearest-point subroutine to the "ranked" versions of each point.

Similarly, it's also possible to get maximize Pearson correlation by pre-normalizing all your points down to mean $0$ and constant variance.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.