Author: Dann Corbit
Date: 12:06:21 10/13/04
Go up one level in this thread
On October 13, 2004 at 09:08:19, Norm Pollock wrote: >On October 12, 2004 at 16:29:52, Dann Corbit wrote: > >>On October 12, 2004 at 14:50:51, Jorge Pichard wrote: >> >>>Petir 1.7 was using the WizardLight8 Opening book :-) >>> >>>Rank Engine Score Uf Pe S-B >>>1 Petir 1.7 11.0/20 · ·· ·· 00011=01=1011=01110= 99.00 >>>2 Ufim501 9.0/20 11100=10=0100=10001=·· ·· · 99.00 >>> >>> >>>20 games played / Tournament finished >>>Tournament start: 2004.10.10, 20:52:20 >>>Latest update: 2004.10.12, 18:41:28 >>>Site/ Country: S0025128408, United States >>>Level: Blitz 15/1 >>>Hardware: AMD Athlon(tm) Processor 1202 MHz >>>Operating system: Microsoft Windows 2000 Build 2195 >>>PGN-File: C:\Program Files\Arena\Arena.pgn >>>Table created with: Arena 1.092 >> >>If you were to add 10 more games so that your tournaments had 30 games each, the >>statistical significance goes up quite a bit. >> >>30 is a sort of magic number when it comes to statistical measurements. > >I remember the "magic number" as 32. It sort of allows the use of a normal >distribution to approximate a binomial distribution. Something like that. Well, if 30 is good, then 32 would be fractionally better, of course. From here: http://www.proaxis.com/~johnbell/sfpp/sfpp6.htm We have this: "Luckily, some nice person has figured this out, and published another table called the "t-table". It is very close to the Z-table except in the very small sample sizes. In fact, after a sample size of about 30 or so there is virtually no difference (which just means that you are now getting a very good estimate of the standard deviate). You often hear in statistics that "after a sample size of 30 it is correct to use the Z-table". This isn't really true, but there is so little difference between the tables that nobody worries about doing it." From here: http://www.jsc.nasa.gov/bu2/PCEH/PCEH-D.DOC We have this: "There is one complication with this result. To be able to judge the size of the error we might make when we use as an estimate of , we must know the value of the population standard deviation, . Since this is not the case in most practical situations, we have no choice but to replace with an estimate, usually the sample standard deviation, s. In general, this is considered to be reasonable provided the sample is sufficiently large (n >= 30)." Other citations for 30: http://home.xnet.com/~fidler/triton/math/review/mat170/ssmean/ssmean1.htm http://qudata.com/online/statcalc/ http://arnoldkling.com/apstats/ttests.html http://www.bized.ac.uk/timeweb/digging/dig_source_illus.htm http://instructional1.calstatela.edu/mfinney/Courses/309/note.html At any rate, good things start to happen when you have 30 or more observations. And when you have significantly less than 30 observations, you should use the T test tables (or perform the integration yourself).
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.