Table 5:

Simulation results from 10,000 MC runs when the true model is a near upper hinge logistic model and the data are fit using an upper hinge model. Mean estimate, %bias, and MC SD of the parameter estimate are given. Additionally, coverage and average standard error (SE) are given for model-based and model-robust standard error estimates.

				Model-based		Model-robust
	n	Est(%Bias)	MC SD	Coverage	SE	Coverage	SE
${\mathit{\beta }}_{\mathit{z}}$
	50	1.28 (29)	0.57	0.986	0.42	0.959	0.63
	100	1.11 (12)	0.27	0.969	0.23	0.958	0.30
	500	1.02 (2)	0.09	0.953	0.09	0.95	0.09
	2000	1.00 (1)	0.05	0.947	0.05	0.946	0.05
${\beta }_{(x-e)-}$
	50	1.92 (95)	1.84	0.975	1.78	0.933	6.59
	100	1.45 (47)	1.22	0.954	0.74	0.953	3.83
	500	1.03 (4)	0.23	0.909	0.18	0.984	0.39
	2000	1.00 (1)	0.10	0.915	0.09	0.984	0.13
$e$
	50	3.01	1.31	0.914	1.53	0.788	7.81
	100	3.02	1.08	0.905	0.99	0.915	5.08
	500	3.03	0.47	0.919	0.42	0.992	0.99
	2000	3.03	0.22	0.922	0.21	0.987	0.37