Table 6.

Properties and performance of the linear quadrature, optimal in the minimax sense, compared to the optimal trapezoid and GL or CC approaches for Model 5

$n$	$c_v[\%]$	Method	$\sqrt{\mathrm{Objective}}$	Bias	RMSRE	Maximum deviance	Maximum relative deviance
6	10	optimal	4.40E−2	−0.202	4.22E−2	−28.3	.1559
		opt. trap.	4.52E−2	−1.567	4.45E−2	−31.1	−.1635
		GL	5.54E−2	−0.030	4.85E−2	30.3	.1881
	5	optimal	2.24E−2	−0.251	2.16E−2	−13.5	−.0876
		opt. trap.	2.49E−2	−1.229	2.36E−2	−15.2	−.0956
		GL	2.77E−2	−0.019	2.40E−2	15.5	.1110
	0	optimal	4.67E−5	−0.0004	2.21E−5	9.43E−2	4.65E−5
		opt. trap.	1.47E−2	−1.3228	1.05E−2	−2.827	1.47E−2
		GL	5.03E−4	0.0133	1.45E−4	0.026	4.90E−4
4	10	optimal	5.39E−2	0.325	5.14E−2	−39.2	.2087
		opt. trap.	6.00E−2	−2.808	5.80E−2	−38.9	−.2236
		GL	6.64E−2	0.357	5.73E−2	−38.0	.2132
	5	optimal	2.73E−2	0.641	2.65E−2	19.1	.1082
		opt. trap.	3.69E−2	−2.358	3.32E−2	−21.6	−.1324
		GL	3.54E−2	0.433	2.92E−2	−22.3	.1142
	0	optimal	6.87E−4	−0.021	3.76E−4	0.14	6.85E−4
		opt. trap.	2.61E−2	−2.271	1.80E−2	−5.26	−2.60E−2
		GL	1.42E−2	0.466	4.90E−3	0.75	1.40E−2
2	10	optimal	0.078	0.64	0.074	−50.5	.3123
		opt. trap.	0.144	−11.78	0.120	−66.4	.4163
		GL	0.203	10.42	0.123	58.9	.4326
	5	optimal	0.045	1.37	0.039	30.3	.1656
		opt. trap.	0.128	−11.87	0.101	−48.9	−.2419
		GL	0.180	10.39	0.099	40.0	.2712
	0	optimal	4.02E−3	0.12	2.17E−3	−0.88	.0040
		opt. trap.	1.23E−1	−11.96	9.41E−2	−22.93	−.1220
		GL	1.72E−1	10.41	8.96E−2	12.10	.1633