GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-02 > 1266678533


From: "Ken Nordtvedt" <>
Subject: [DNA] Clade Age Sigmas
Date: Sat, 20 Feb 2010 08:08:53 -0700


Statistical 2-sigma confidence intervals for clade age estimates should be about a 1000 years for the conditions stated below. Showing half this size is unrealistic:

Age itself of about 3900 years
Sample size of about 100
Total y haplotype mutation rate M of about 1/14
Sample haplotype tree of assumed constant growth rate from 1 to N.

If you want to adjust to different parameters and tree structure than used above then use: 1-Sigma is proportional to SqRoot { .7 G / M logN }
log is natural log.

If you have reason to believe your haplotype samples' tree has its nodes earlier then sigma improves; if the tree has its nodes later then sigma gets worse.

Ken


This thread: