## And now it works [R for BE/BA]

I think I made beta work now, thanks a lot PharmCat

It appears that beta can be constructed even if the local Xi is not of rank p; I guess it must be a little typo in Gurka's paper. Anyways, I reprogrammed the lines, and now I get conformable matrices, and beta is the same whichever way I calculate it.
May try and see if I can do some comparisons about the performance, I still need a lot of code to put it all into function.

Thanks PharmCat, pointing me towards Gurka and insisting on Xi was a great help.

Pass or fail!
ElMaestro