Table 5. Simulation results with...

Table 5.

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

Case	Estimator	MSE	Bias	Variance	Coverage	Length
RE	OLS	0.19	3.43	0.17	0.93	0.16
	GMM	0.21	3.66	0.08	0.77	0.11
	GMM-GLS-IV	0.14	2.96	0.13	0.94	0.14
	GLS-IV	0.15	2.99	0.12	0.93	0.13
General	OLS	3.66	18.05	0.37	0.19	0.24
	GMM	0.79	6.98	0.31	0.75	0.22
	GMM-GLS-IV	0.09	2.39	0.11	0.96	0.12
	GLS-IV	0.10	2.42	0.11	0.95	0.12

Case	Estimator	MSE	Bias	Variance	Coverage	Length
RE	OLS	0.19	3.43	0.17	0.93	0.16
	GMM	0.21	3.66	0.08	0.77	0.11
	GMM-GLS-IV	0.14	2.96	0.13	0.94	0.14
	GLS-IV	0.15	2.99	0.12	0.93	0.13
General	OLS	3.66	18.05	0.37	0.19	0.24
	GMM	0.79	6.98	0.31	0.75	0.22
	GMM-GLS-IV	0.09	2.39	0.11	0.96	0.12
	GLS-IV	0.10	2.42	0.11	0.95	0.12

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