Table 3.

Results from the main simulation study with a correlation matrix estimated in an independent sample of 10,000 individuals

Parameter	Method	Pruning	Mean	SD	Mean SE	Power
${\theta }_1$	MV‐IVW	0.4	0.303	0.166	0.147	57.1
		0.6	0.206	0.114	0.090	62.4
	MV‐LIML	0.4	0.340	0.189	0.162	59.0
		0.6	0.301	0.183	0.098	76.7
	MV‐IVW‐PCA	–	0.295	0.130	0.118	69.2
	MV‐LIML‐PCA	–	0.347	0.153	0.130	74.7
${\theta }_2$	MV‐IVW	0.4	−0.013	0.167	0.148	7.3
		0.6	−0.005	0.114	0.090	11.2
	MV‐LIML	0.4	−0.009	0.193	0.163	7.3
		0.6	0.003	0.174	0.098	20.8
	MV‐IVW‐PCA	–	−0.012	0.129	0.118	7.2
	MV‐LIML‐PCA	–	−0.006	0.154	0.130	8.9
${\theta }_3$	MV‐IVW	0.4	−0.485	0.165	0.148	86.8
		0.6	−0.322	0.126	0.090	88.9
	MV‐LIML	0.4	−0.528	0.188	0.164	87.9
		0.6	−0.436	0.207	0.099	93.6
	MV‐IVW‐PCA	–	−0.476	0.130	0.119	96.1
	MV‐LIML‐PCA	–	−0.538	0.152	0.131	97.0

Note: Mean estimates, standard deviation (SD) of estimates, mean standard error (mean SE) of estimates, and empirical power of the 95% confidence interval to estimate ${\theta }_1=0.4$, ${\theta }_2=0$, and ${\theta }_3=-0.6$.