Table 6. Simulation results with...

Table 6.

Simulation results with non-exogenous instruments. RE implies MA(3) errors; general implies ARMA(1,3) errors

Case	Estimator	MSE	Bias	Variance	Coverage	Length
RE	OLS	0.32	4.63	0.25	0.89	0.19
	GMM	0.37	4.84	0.19	0.82	0.17
	GMM-GLS-IV	0.17	3.28	0.13	0.92	0.14
	GLS-IV	0.23	3.74	0.13	0.87	0.14
General	OLS	3.96	18.46	0.47	0.26	0.26
	GMM	1.67	10.63	0.56	0.65	0.26
	GMM-GLS-IV	0.19	3.47	0.15	0.91	0.15
	GLS-IV	0.26	3.71	0.15	0.89	0.14

Case	Estimator	MSE	Bias	Variance	Coverage	Length
RE	OLS	0.32	4.63	0.25	0.89	0.19
	GMM	0.37	4.84	0.19	0.82	0.17
	GMM-GLS-IV	0.17	3.28	0.13	0.92	0.14
	GLS-IV	0.23	3.74	0.13	0.87	0.14
General	OLS	3.96	18.46	0.47	0.26	0.26
	GMM	1.67	10.63	0.56	0.65	0.26
	GMM-GLS-IV	0.19	3.47	0.15	0.91	0.15
	GLS-IV	0.26	3.71	0.15	0.89	0.14

Note(s):

Weekly data for US-CAD, US-JP for the period November 2010 to April; 2020 (⁠ $T = 492$ ⁠). For OLS we use HAC standard errors as described in the text

[ViewLarge]

Sharing Unavailable