GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2011-02 > 1297180929


From: "Ken Nordtvedt" <>
Subject: [DNA] variance quiz game
Date: Tue, 8 Feb 2011 09:02:09 -0700


Suppose you have two haplotypes and want its estimated tmrca from variance. They have only 4 markers ("Oppenheimer haplotypes"), and the markers have identical mutation rates m. Standard procedure would be to use the estimator:

G = [ Var(1) + Var(2) + Var(3) + Var(4) ] / 8m

But here is another estimator:

G^2 = [ Var(1)Var(2) + Var(1)Var(3) + Var(1)Var(4) + Var(2)Var(3) + Var(2)Var(4) + Var(3)Var(4) ] / 24m^2

Is one of these estimators better than other and why?

Var(i) in the above is the ith marker's repeat number difference squared between haplotypes.


This thread: