While I'm in favour of robust methods like Farr's law, one question the article doesn't mention is: what would the law's prediction have been had it been applie

While I'm in favour of robust methods like Farr's law, one question the article doesn't mention is: what would the law's prediction have been had it been applied ten years ago? This gives us some idea of the stability of the result. The paper (open access for the win!) is here: http://www.injepijournal.com/content/1/1/31 http://www.injepijournal.com/content/1/1/31 It doesn't seem to have run the alternative-date analyses, which is too bad.