I am looking for a data application to test a particular regression algorithm.
In particular, I am looking for an application where the data items are noisy samples coming from a normalised function f(x), that we don't observe. By normalised I mean that the integral of f(x) in its support is ∫f(x)dx = 1. The algorithm attempts to reconstruct the normalised f(x) given the observed samples that are contaminated with noise.
I would very much appreciate it if somebody could point me to such a dataset, or perhaps to a domain where this problem naturally appears.