Table 1.

Statistical properties of different methods as applied to the analysis of data generated according to the model (15)

n	Model	Power1	Power2	Power3	FP	FDR
200	fixed	0.86	0.92	0.06	4.38	0.69
		[2.99, 0.18]	[3.26, 0.30]	[3.83, 0.08]
	fixed	0.88	0.85	0.26	3.67	0.63
	PC = 3	[2.55, 0.13]	[2.95, 0.24]	[2.80, 0.10]
	fixed	0.91	0.56	0.46	0.5	0.17
	PC = 10	[1.49, 0.09]	[1.86, 0.13]	[1.92, 0.13]
	mixed	0.89	0.78	0.78	0.04	0.01
	$\mu \ne 0$	[1.13, 0.06]	[1.92, 0.08]	[1.99, 0.08]
	mixed	0.71	0.52	0.04	0.01	0.00
	μ = 0	[0.46, 0.05]	[0.87, 0.06]	[0.85, 0.06]
	mixed	0.71	0.45	0.10	0.02	0.01
	PC = 3	[1.14, 0.12]	[1.60, 0.22]	[1.10, 0.17]
	mix	0.77	0.34	0.28	0.04	0.02
	PC = 10	[1.17, 0.09]	[1.40, 0.17]	[1.28, 0.17]
400	fixed	1	1	0.36	11.0	0.81
		[2.62, 0.12]	[1.91, 0.23]	[1.53, 0.18]
	fixed	0.99	0.95	0.57	6.73	0.71
	PC = 3	[1.49, 0.10]	[1.65, 0.17]	[1.31, 0.12]
	fixed	1.00	0.92	0.93	0.83	0.19
	PC = 10	[0.68, 0.07]	[1.06, 0.08]	[1.00, 0.08]
	mixed	0.99	0.98	0.98	0.05	0.01
	$\mu \ne 0$	[0.46, 0.05]	[0.87, 0.06]	[0.85, 0.06]
	mixed	0.98	0.86	0.26	0.01	0.00
	μ = 0	[0.52, 0.08]	[0.86, 0.12]	[0.54, 0.04]
	mixed	0.98	0.80	0.41	0.02	0.01
	PC = 3	[0.50, 0.07]	[0.84, 0.11]	[0.67, 0.6]
	mixed	0.99	0.81	0.82	0.06	0.02
	PC = 10	[0.51, 0.07]	[0.79, 0.07]	[0.77, 0.07]

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.