Table 4.

Statistical properties of different methods as applied to the analysis of data generated according to the model (15) but with Laplace distribution of the polygenic variables

n	Model	Power1	Power2	Power3	FP	FDR
200	fixed	0.865	0.931	0.073	5.84	0.742
		[3.29,0.16]	[3.5,0.29]	[2.4,0.11]
	fixed	0.898	0.829	0.272	4.027	0.651
	PC = 3	[2.56,0.13]	[3.03,0.23]	[2.51,0.12]
	fixed	0.921	0.577	0.498	0.736	0.233
	PC = 10	[1.64,0.09]	[2.22,0.12]	[2,0.11]
	mixed	0.883	0.741	0.753	0.044	0.015
	$\mu \ne 0$	[1.28,0.07]	[2.09,0.08]	[1.95,0.07]
	mixed	0.652	0.451	0.027	0.013	0.008
	μ = 0	[1.11,0.13]	[1.75,0.22]	[0.81,0.18]
	mixed	0.653	0.393	0.083	0.021	0.012
	PC = 3	[1.11,0.12]	[1.73,0.22]	[1.3,0.17]
	mixed	0.756	0.284	0.233	0.036	0.019
	PC = 10	[1.23,0.09]	[1.71,0.17]	[1.45,0.16]
400	fixed	0.992	0.992	0.347	11.32	0.819
		[2.2,0.12]	[2.06,0.21]	[1.71,0.18]
	fixed	0.991	0.951	0.593	7.384	0.728
	PC = 3	[1.58,0.1]	[1.59,0.16]	[1.27,0.12]
	fixed	0.999	0.943	0.94	1.179	0.249
	PC = 10	[0.79,0.07]	[1.05,0.08]	[1.04,0.08]
	mixed	1	0.987	0.986	0.048	0.012
	$\mu \ne 0$	[0.56,0.05]	[0.88,0.06]	[0.92,0.06]
	mixed	0.982	0.882	0.251	0.006	0.002
	μ = 0	[0.61,0.07]	[0.87,0.11]	[0.52,0.04]
	mixed	0.983	0.803	0.411	0.026	0.008
	PC = 3	[0.58,0.07]	[0.85,0.11]	[0.63,0.06]
	mixed	0.998	0.814	0.815	0.057	0.015
	PC = 10	[0.57,0.06]	[0.83,0.07]	[0.82,0.07]

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.