Author: Ed Panek
Date: 09:44:43 07/09/03
Go up one level in this thread
Obviously as the number of games increases, the chance of the sample mean and the Popluation mean differing decreases with each game played. The formula for samples over 30 is given as: (X-u)/(S/SQRT(n)) (where X is the sampled mean, u is the true mean and S is the variance of the data)if everything in this formula stays constant and only n(the sample size)increases, the denominator S/SQRT(N) gets smaller. When the denominator gets smaller the overall fraction gets larger. When the fraction gets larger, the Z score may go from a 1 to a 2 safety range granting greater freedom under the normal distribution curve... meaning that 95% certainty becomes 99% certainty. That implies that the chance that the hypothesized mean is actually the the true mean is 99% certain as long as the distribution is standard bell shaped. There is no 100% certainty without sampling the entire population. Ed
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.