I just came here from your LinkedIn post. Just to be clear - is the error you describe in your LI post as a 'gross arithmetical error' the loss of precision after the 12th decimal, in cases when there is not a format applied to the result? If so, this just looks like a floating point issue to me.
I'm just trying to understand whether I'm missing something obvious, as I'm not a statistician.
Nik
... View more