Skip to main content
. Author manuscript; available in PMC: 2009 Nov 25.
Published in final edited form as: Ann Hum Genet. 2009 Mar 25;73(Pt 3):346–359. doi: 10.1111/j.1469-1809.2009.00515.x

Table V.

Performance of the two-candidate-haplotype log-linear model enforcing HWE and R2=R12 and T2=T12 for Yin-Yang haplotypes. Genotypes were simulated mimicking RFC1 haplotypes under HWE with “Yin” as the risk-relevant haplotype, conferring either risk or protection. Each scenario reflects 5000 simulations with 400 triads in each simulation.

Ratio of
frequencies
(Yin:Yang)		 Pattern
of
Missing
data 1 		Simulated
R1 (Yin
haplotype)		 Power
(1df Yin)		 Power
(2df
Yin+Yang)		 Empirical
coverage of
nominal 95%
confidence
region (2 risk
parameters)		 Geometric mean of Estimated R1 of risk
haplotype (95% CI)2
 Error
Rate3
Yin Yang
1:1
(0.141:0.141)		 None 1.414 0.730 0.631 0.954 1.42(1.41,1.43) 1.00(1.00,1.01) 0.040
0.707 0.617 0.506 0.950 0.71(0.70,0.71) 1.00(0.99,1.00) 0.043
20% g 1.414 0.686 0.585 0.956 1.42(1.42,1.43) 1.00(1.00,1.01) 0.039
0.707 0.574 0.467 0.948 0.71(0.70,0.71) 1.00(0.99,1.00) 0.042
1:3
(0.071:0.212)		 None 1.414 0.479 0.373 0.949 1.42(1.41,1.43) 1.00(1.00,1.01) 0.036
0.707 0.385 0.297 0.949 0.70(0.70,0.71) 1.00(1.00,1.00) 0.033
20% g 1.414 0.430 0.337 0.950 1.42(1.41,1.43) 1.00(1.00,1.01) 0.039
0.707 0.344 0.261 0.950 0.70(0.70,0.71) 1.00(1.00,1.00) 0.037
1

None: no missing; 20% g: 20% of SNP genotypes are missing randomly.

2

The 95% CI is based on the empirical standard error calculated using the estimates from 5000 independent simulations.

3

Error rate incurred by designating the haplotype whose log relative risk deviated most from 0 as the risk-relevant one.