Compare cumulative distribution functions of several function point counters Function point counters are used to measure the size of software. However, they are believed not to deliver exact results. Different measurements sometimes produce different results.
The performance of function point counters is described by the cumulative distribution function of the number of function points measured by that counter.
Fi(x) gives the probability that the number of function points counted by the counter i is less than or equal to x. The variable i describes the number of the current counter (in my study, 17 different counters are present).
I would like to test the hypothesis that Fi(x) and Fj(x) are equal for all i and j, i ≠ j for the 17 analyzed function point counters. Which statistical method could be used for that?
 A: You should maybe give some more relevant details, like sample sizes and how the samples were obtained, so that we can judge if you have an independent sample.
What you seems to have, is a sample of 17 empirical distribution functions, and you want to test the null hypothesis that the 17 underlying theoretical distribution functions all are the same.  If you only had two empirical distribution functions, you could have used the Kolmogorov-Smirnof test, or some alternative to that.  But since you have 17, it seems you need some kind of ANOVA for empirical distribution functions. 
Here:  http://projecteuclid.org/euclid.aoms/1177706261   is a paper from 1959 about a k-sample analog of the Kolmogorov-Smirnoff test.  And here:  http://www.jstor.org/discover/10.2307/2238210?sid=21105703618263&uid=2&uid=3738744&uid=4   is a paper from 1965 about same topic.  Hope this can give you a starting point!
Another idea is to look into functional data analysis, search this site, there are many interesting posts!
