[Archive]

Posted By: Glenn

Frank,

I don't think there's any reason to have raised the issue differently. I actually think it's a very interesting and perfectly valid question to raise, and I'm glad to have a thread I can finally contribute something to. However, whatever the case may be (whether there is a difference between the two series or not) we can't appropriately resolve the issue through a survey of self-selected participants. Perhaps I'm wrong, but I think any other statisticians here would agree with me on that. In any case, I wasn't trying to refute you, I was just trying to evaluate the relevant data in the way I've been trained, so that we could see if your suspicion was correct.