Gaussian Copula Regression Modeling for Marker Classification Metrics with Competing Risk Outcomes

Table 1

Empirical bias, standard error (SE), and mean square error (MSE) of the estimators of the parametric Gaussian copula regression model for simulations generated from a Gaussian copula, with sample sizes 200 and 400 and censoring distributions and .

Sample size and censoring	Parameters	Bias	SE	MSE
censoring		−0.0078	0.9158	0.8387
		−0.0030	0.0878	0.0077
		0.0094	0.0977	0.0096
		−0.1236	0.6693	0.4632
		−0.0009	0.0627	0.0039
		−0.0928	0.4011	0.1695
		−0.0250	0.0822	0.0074
		0.0270	0.2416	0.0591
		−0.0393	0.2946	0.0883
		−0.0101	0.0772	0.0061
		0.0234	0.0565	0.0037
		0.0013	0.1080	0.0117
censoring		−0.0634	0.6162	0.3838
		0.0028	0.0593	0.0035
		0.0083	0.0718	0.0052
		−0.0394	0.4825	0.2344
		0.0066	0.0393	0.0016
		−0.0431	0.2895	0.0857
		−0.0176	0.0638	0.0044
		0.0143	0.1670	0.0281
		−0.0024	0.2003	0.0401
		0.0017	0.0488	0.0024
		0.0209	0.0472	0.0027
		−0.0066	0.0668	0.0045
censoring		−0.0731	0.9864	0.9783
		−0.0025	0.0943	0.0089
		0.0098	0.1173	0.0139
		−0.1441	0.8617	0.7633
		−0.0006	0.0660	0.0044
		−0.1841	0.4170	0.2078
		−0.0544	0.0881	0.0107
		−0.0143	0.2482	0.0618
		0.0069	0.3008	0.0905
		0.0021	0.0742	0.0055
		0.0177	0.0653	0.0046
		−0.0060	0.1032	0.0107
censoring		0.1473	0.7044	0.5178
		−0.0214	0.0689	0.0052
		−0.0081	0.0773	0.0060
		−0.0741	0.5356	0.2924
		0.0083	0.0441	0.0020
		−0.2557	0.3006	0.1558
		−0.0746	0.0635	0.0096
		−0.0260	0.1569	0.0253
		0.0161	0.1885	0.0358
		−0.0012	0.0571	0.0033
		0.0184	0.0473	0.0026
		−0.0088	0.0710	0.0051