c++ statistics code

I am wondering if there is some repository of c++ code for statistics comparing two columns of numbers? Statistics are so common now that it seems like with the oo concept of reusable code there should be allot of that floating around. Code that would accept two vectors and generate comparisons like pearson's correlation, root mean square error, mean absolute error, median absolute error, etc would be very useful.

I have looked around some and find some code examples, and other posts basically saying, "Google the algorithm and figure it out yourself." At this point, there should almost be stl classes like the ones for union(), intersection(), accumulate(), and other simpler math functions.

Is there anything like this out there that someone can point me to?

LMHmedchem

Wikipedia has a list of numerical libraries with a C++ section that you can check out.

take a look at the boost accumulator library ( and the boost math library for statistics oriented math support ). Moreover, I'm not a fan neither of scripting languages nor of "general purpose" statistics ( because people systematically abuse them ) but statsitics oriented languages like R ( or python ) seems quite popular among scientists, so it may be worth considering to have bindings with your c++ program to them if performance is not a primary requirement.

Thanks for the suggestions. I ended up writing my own code based on available examples. There are a number of available libraries that support these calculations, but the functions weren't very hard to write. I had to make some adjustments because I calculate the statistics on sub-sets of my vectors, so I probably couldn't have used anything canned anyway.

I will post my code when I get a chance in case anyone else is interested.

LMHmedchem