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

BGonline.org Forums

Fast Rollouts

Posted By: Timothy Chow
Date: Sunday, 6 December 2009, at 9:09 p.m.

In Response To: Fast Rollouts (Bob Koca)

Often, if you have a probability distribution that is highly skewed—say it has exponentially low probability of attaining certain exponentially high values—then if you try to estimate the variance of the distribution by randomly sampling from it and calculating the sample variance, your estimate will (with high probability) be a serious underestimate unless you take exponentially many samples.

The doubling cube can sometimes cause this kind of skewing. I would, however, be surprised if variance estimates at DMP were consistently found to be low.

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.