[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

A study: How confident can we be of the confidence interval!

Posted By: eXtreme Gammon
Date: Sunday, 21 November 2010, at 4:50 a.m.

In Response To: A study: How confident can we be of the confidence interval! (Bob Koca)

UBK 13 out of the 14 values differ in the same direction. How to explain that? Did you do any formal hypothesis test?

I assume it has to do with the assumption that the distribution is normal around the average. If i get a chance i'll graph the distribution of these 1068 samples to compare it to the normal distribution graph. That will show us (for that particular position) if it's due to a skewness of the distribution or if the stdev is just smaller than the normal assumption says.

UBK I would be interested in a similar study but looking at the difference between two plays and seeing if independence assumption is warranted. I think it often will not be.

Yes that will be interesting to check, if you have a specific position you like to use, please post here or email me. I can rollout each position N time and compute for each game the delta of the equity. If the average and stdev of the difference game per game match the Normal assumption then will have a better idea of the validity of the independence assumption (as I type, I realize the average will match, but the stdev will be interesting to see)

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.