Table 5.

Statistical properties of different methods as applied to the analysis of data generated according to the model (17), with no polygenic background

n	Model	Power1	Power2	Power3	FP	FDR
200	fixed	0.80	0.80	0.82	0.18	0.06
		[1.83, 0.05]	[2.91, 0.05]	[3.37, 0.06]
	fixed	0.73	0.59	0.60	0.14	0.07
	PC = 3	[1.99, 0.06]	[2.80, 0.07]	[3.02, 0.07]
	fixed	0.59	0.20	0.22	0.1	0.09
	PC = 10	[1.70, 0.07]	[2.01, 0.17]	[2.54, 0.18]
	mixed	0.62	0.46	0.49	0.04	0.02
	$\mu \ne 0$	[1.66, 0.06]	[2.53, 0.07]	[2.83, 0.07]
	mixed	0.63	0.47	0.48	0.03	0.02
	μ = 0	[1.56, 0.67]	[2.41, 0.07]	[2.75, 0.07]
	mixed	0.57	0.34	0.35	0.03	0.02
	PC = 3	[1.57, 0.07]	[2.38, 0.10]	[2.72, 0.10]
	mixed	0.50	0.14	0.16	0.05	0.06
	PC = 10	[1.46, 0.10]	[1.75, 0.21]	[2.53, 0.22]
400	fixed	1.00	0.99	0.98	0.08	0.02
		[0.96, 0.05]	[1.72, 0.04]	[1.73, 0.04]
	fixed	1.00	0.96	0.97	0.05	0.02
	PC = 3	[1.09, 0.04]	[1.68, 0.05]	[1.61, 0.05]
	fixed	0.98	0.66	0.67	0.06	0.02
	PC = 10	[1.10, 0.05]	[1.46, 0.06]	[1.35, 0.06]
	mixed	0.98	0.95	0.95	0.02	0.01
	$\mu \ne 0$	[1.02, 0.04]	[1.72, 0.04]	[1.72, 0.04]
	mixed	0.99	0.95	0.95	0.02	0.01
	μ = 0	[1.01, 0.04]	[1.70, 0.04]	[1.66, 0.04]
	mixed	0.93	0.87	0.87	0.03	0.01
	PC = 3	[1.03, 0.04]	[1.59, 0.05]	[1.58, 0.04]
	mixed	0.93	0.56	0.55	0.02	0.01
	PC = 10	[1.00, 0.04]	[1.34, 0.08]	[1.31, 0.08]

We report Power, average number of false positives (FP), and false discovery rate (FDR). The square brackets report a mean distance to a simulated QTL and a mean value of $|{\widehat{\beta }}_i-{\beta }_i|$. A fixed effects model and a mixed model with μ = 0 are additionally supplemented with the first 3 or 10 PCs of the incidence matrix X.