Table 3. Results of logistic regression...

Table 3.

Results of logistic regression and system GMM with alternative forex exposure specifications

Panel A. Logistic regressions						Panel B. System GMM
Variables	Model I	Model II	Model III	Model IV	Model V	Variables	Model I	Model II	Model IV
FS Ratio	0.025 *** (0.002) 1.025					FS Ratio	0.00004*** (0.000)
LnTFCF		0.197*** (0.017) 1.218				LnTFCF		0.00031*** (0.000)
TFCF/TA			2.244*** (0.322) 9.435
LnNFCF				0.173*** (0.015) 1.189		LnNFCF			0.00026*** (0.000)
NFCF/TA					3.013*** (0.556) 20.344	NFCF/TA
						NV/TA(−1)	0.702*** (0.005)	0.677*** (0.006)	0.686*** (0.006)
Diagnostic tests:
Hosmer–Lemeshow	0.120	0.312	0.158	0.279	0.154	No. of Instruments	95	95	95
AUC-ROC curve	0.8172	0.8236	0.7811	0.8159	0.7769	No. of Groups	250	250	250
Pseudo R²	0.2275	0.2469	0.1877	0.2332	0.1805	AR2	0.769	0.771	0.770
Likelihood Ratio (p-value)	0.0000	0.0000	0.0000	0.0000	0.0000	Hansen test	0.671	0.821	0.717
Likelihood Ratio (p-value)						Diff-in-Hansen test	0.645	0.682	0.689
						Wald test	0.000	0.000	0.000

Panel A. Logistic regressions						Panel B. System GMM
Variables	Model I	Model II	Model III	Model IV	Model V	Variables	Model I	Model II	Model IV
FS Ratio	0.025 ***(0.002)1.025					FS Ratio	0.00004***(0.000)
LnTFCF		0.197***(0.017)1.218				LnTFCF		0.00031***(0.000)
TFCF/TA			2.244***(0.322)9.435
LnNFCF				0.173***(0.015)1.189		LnNFCF			0.00026***(0.000)
NFCF/TA					3.013***(0.556)20.344	NFCF/TA
						NV/TA(−1)	0.702***(0.005)	0.677***(0.006)	0.686***(0.006)
Diagnostic tests:
Hosmer–Lemeshow	0.120	0.312	0.158	0.279	0.154	No. of Instruments	95	95	95
AUC-ROC curve	0.8172	0.8236	0.7811	0.8159	0.7769	No. of Groups	250	250	250
Pseudo R²	0.2275	0.2469	0.1877	0.2332	0.1805	AR2	0.769	0.771	0.770
Likelihood Ratio (p-value)	0.0000	0.0000	0.0000	0.0000	0.0000	Hansen test	0.671	0.821	0.717
Likelihood Ratio (p-value)						Diff-in-Hansen test	0.645	0.682	0.689
						Wald test	0.000	0.000	0.000

Notes:

For definitions of variables, refer to Table 1. In all models, N = 2,250 firm-year observations. In Panel A, the dependent variable is a binary variable HS, and figures in each cell correspond to the coefficient value, standard errors in parenthesis and odds ratio. In Panel B, the dependent variable is the notional value ratio (NV/TA), and figures in each cell correspond to the coefficients and standard errors in parenthesis. All estimated models include Shari'ah and all control variables, but the results are not reported since they are consistent with Table 4. AUC-ROC denotes Area Under the Receiver Operating Characteristic Curve. Asterisks ***, ** and * indicate significant at 1, 5 and 10% levels, respectively

[ViewLarge]

Sharing Unavailable