|
BGonline.org Forums
Got Variance?
Posted By: Timothy Chow In Response To: Got Variance? (Daniel Murphy)
Date: Sunday, 23 May 2010, at 10:41 p.m.
Well, I'm still not clear on what people are claiming (and different people may be claiming different things). If the claim is that in many typical positions, the reported confidence interval is an underestimate, then picking a particular contrived example like yours won't prove anything one way or the other. One would need to take a particular position where people think the confidence interval is not what it seems to be and "roll it out to death" as Bob suggested.
One problem here is figuring out what "rolling it out to death" really means. For example I doubt that 46K is anywhere near enough for these purposes. Supposing that we determine that several million trials are needed, I doubt that anyone will be willing to devote the computer resources.
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.