Table 2.

Statistical properties of different methods as applied to the analysis of data generated according to the model (15) when the markers are spaced every 5 cM.

n	Model	Power1	Power2	Power3	FP	FDR
200	fixed	0.92	0.94	0.11	6.58	0.76
		[2.68,0.16]	[4.04,0.27]	[3.67,0.13]
	fixed	0.93	0.85	0.31	4.52	0.66
	PC = 3	[2.07,0.12]	[3.83,0.2]	[3.51,0.1]
	fixed	0.92	0.55	0.48	0.74	0.24
	PC = 10	[0.93,0.09]	[3.22,0.1]	[3.17,0.09]
	mixed	0.91	0.80	0.80	0.08	0.02
	$\mu \ne 0$	[0.67,0.06]	[3.17,0.07]	[3.29,0.07]
	mixed	0.59	0.37	0.01	0.004	0.002
	μ = 0	[0.59,0.12]	[2.95,0.19]	[2.46,0.11]
	mixed	0.64	0.37	0.07	0.03	0.02
	PC = 3	[0.72,0.11]	[3.03,0.18]	[2.96,0.13]
	mixed	0.73	0.29	0.23	0.07	0.03
	PC = 10	[0.72,0.09]	[2.86,0.13]	[2.83,0.13]
400	fixed	1	0.99	0.39	11.98	0.82
		[1.63,0.12]	[3.37,0.18]	[2.94,0.19]
	fixed	0.999	0.95	0.59	7.70	0.73
	PC = 3	[1,0.1]	[3.16,0.14]	[2.77,0.12]
	fixed	1	0.89	0.90	1.03	0.23
	PC = 10	[0.41,0.07]	[2.7,0.08]	[2.68,0.08]
	mixed	0.999	0.98	0.98	0.10	0.02
	$\mu \ne 0$	[0.17,0.06]	[2.61,0.07]	[2.68,0.06]
	mixed	0.93	0.65	0.07	0.006	0.002
	μ = 0	[0.14,0.07]	[2.51,0.08]	[2.32,0.04]
	mixed	0.95	0.65	0.24	0.03	0.008
	PC = 3	[0.13,0.07]	[2.59,0.08]	[2.43,0.05]
	mixed	0.98	0.66	0.68	0.08	0.02
	PC = 10	[0.14,0.06]	[2.47,0.06]	[2.54,0.06]

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. The polygenic effects are uniformly distributed at the distance of 1 cM but the search is performed over markers spaced 5 cM.