Table 2.

Difference in expected genetic gain from using a Gaussian kernel with respect to a GBLUP prediction model, for low and high panel densities.

			Genetic gain difference			Hypothesis tests
Markers	Species	Trait	Mean	Lower	Upper	sd	eq	noi	Sup
High	Maize	Days to tassel	0.017	0.013	0.021	*		*	*
		Germination count	0.001	–0.003	0.005		*	*
		Number of leaves	0.005	0.000	0.011	*		*
		Plant height	0.018	0.013	0.022	*		*	*
	Rice	Grain length	0.002	–0.003	0.008		*	*
		Grain width	–0.005	–0.008	–0.003	*	*	*
		Grain weight	0.010	0.006	0.013	*		*
		Days to heading	0.002	0.000	0.004		*	*
						0.63	0.5	1	0.25
Low	Maize	Days to tassel	0.037	0.032	0.041	*		*	*
		Germination count	0.004	–0.001	0.008		*	*
		Number of Leaves	0.015	0.009	0.021	*		*
		Plant height	0.018	0.014	0.022	*		*	*
	Rice	Grain length	0.019	0.012	0.025	*		*	*
		Grain width	0.007	0.004	0.010	*		*
		Grain weight	0.019	0.015	0.023	*		*	*
		Days to heading	0.009	0.006	0.012	*		*
	Wheat	Yield 1	0.100	0.084	0.116	*		*	*
		Yield 2	0.105	0.087	0.123	*		*	*
		Yield 3	0.079	0.064	0.094	*		*	*
		Yield 4	0.008	-0.010	0.024			*
						0.83	0.08	1	0.58

Hypothesis tests (sd, statistical differences; eq, equivalence; noi, non-inferiority; sup, superiority), and proportion of models with rejected nulls for high and low panel densities. Asterisks indicate rejected nulls for the corresponding test (see Box 3).