Table 7.

Comparison of bias and efficiency between the proposed pseudo- likelihood approach, the maximum likelihood method [Epstein et al., 2005], and the maximum likelihood approach [Nagelkerke et al. 2004] assuming the Hardy-Weinberg equilibrium and random mating. Each entry lists the mean parameter estimate (standard deviation) for over the 1,000 simulated datasets. Each dataset consists of 500 cases and 500 controls.

			Proposed Method			Epstein			Nagelkerke
p	α	β	triads	case-control	combined	triads	parents	combined	combined
0.1	−1	0	0.006 (0.159)	0.004 (0.161)	0.003 (0.140)	0.007 (0.161)	−0.011 (0.287)	0.003 (0.140)	0.003(0.139)
		.693	0.457 (0.153)	0.699 (0.150)	0.572 (0.127)	0.457 (0.153)	0.851 (0.222)	0.570 (0.130)	0.560(0.124)
	−3	0	−0:001 (0.163)	−0:001 (0.162)	−0:002 (0.140)	0.001 (0.166)	−0:012 (0.298)	−0:002 (0.141)	−0:002(0.138)
		.693	0.660 (0.152)	0.706 (0.159)	0.681 (0.130)	0.660 (0.152)	0.729 (0.240)	0.680 (0.131)	0.676(0.127)
0.2	−1	0	0.005 (0.129)	0.005 (0.134)	0.004 (0.115)	0.005 (0.129)	0.002 (0.245)	0.005 (0.116)	0.004(0.113)
		.693	0.453 (0.130)	0.688 (0.131)	0.568 (0.113)	0.453 (0.130)	0.876 (0.207)	0.564 (0.115)	0.548(0.108)
	−3	0	0.004 (0.126)	−0.001 (0.129)	0.001 (0.109)	0.004 (0.126)	−0.011 (0.242)	0.001 (0.109)	0.004(0.108)
		.693	0.655 (0.130)	0.692 (0.131)	0.673 (0.114)	0.655 (0.130)	0.722 (0.214)	0.673 (0.115)	0.669(0.110)
0.5	−1	0	0.007 (0.136)	−0:003 (0.149)	0.003 (0.126)	0.007 (0.136)	0.001 (0.305)	0.006 (0.126)	0.004(0.120)
		.693	0.456 (0.148)	0.694 (0.152)	0.575 (0.136)	0.456 (0.148)	1.060 (0.353)	0.554 (0.143)	0.527(0.129)
	−3	0	0.005 (0.144)	−0:001 (0.152)	0.003 (0.131)	0.005 (0.144)	−0:003 (0.305)	0.004 (0.131)	0.004(0.124)
		.693	0.661 (0.163)	0.700 (0.168)	0.681 (0.150)	0.661 (0.163)	0.762 (0.386)	0.676 (0.152)	0.670 (0.142)