Table 3

Non-threshold estimated betas (Eq. (1))

DescriptionCQQQFXIGXCMCHI
CoeffSEtCoeffSEtCoeffSEtCoeffSEt
β1.1250.1766.400.8570.1695.060.9170.1665.530.8910.1675.35
Seasonal dummies (Dt):            
January0.0330.0142.420.0150.0170.890.0170.0151.110.0180.0151.17
February0.0010.0140.07−0.0220.014−1.52−0.0100.013−0.80−0.0110.013−0.88
March−0.0230.025−0.93−0.0120.017−0.71−0.0130.017−0.77−0.0120.017−0.71
April−0.0120.016−0.72−0.0050.014−0.39−0.0040.013−0.35−0.0020.014−0.17
May−0.0080.017−0.46−0.0160.010−1.64−0.0130.010−1.29−0.0150.011−1.35
June0.0170.0200.83−0.0080.015−0.520.0030.0150.230.0020.0160.11
July−0.0350.024−1.44−0.0180.023−0.81−0.0190.022−0.88−0.0190.023−0.82
August0.0010.0180.06−0.0120.012−1.06−0.0100.012−0.85−0.0120.013−0.94
September−0.0050.016−0.33−0.0110.012−0.93−0.0070.013−0.57−0.0080.012−0.66
October−0.0040.021−0.18−0.0040.023−0.17−0.0030.020−0.16−0.0060.021−0.29
November0.0070.0170.380.0110.0210.540.0090.0170.550.0100.0190.50
December−0.0140.017−0.86−0.0090.015−0.61−0.0090.012−0.72−0.0050.013−0.41
R20.400  0.335  0.397  0.364  
Adjusted R20.349  0.278  0.345  0.309  
DW1.83  2.35  2.18  2.19  
AIC−2.499  −2.726  −2.900  −2.804  
SIC−2.240  −2.467  −2.641  −2.544  
HQIC−2.394  −2.621  −2.794  −2.698  

Note(s): This table presents the results from estimating Eq. (1) for four Chinese ETFs using monthly data from April 2011 to November 2023. The findings indicate that all estimated beta coefficients are statistically significant at the 1% level. The betas are fixed throughout the sample period, regardless of fluctuations in the EPU index; however, this assumption is indefensible based on the three model selection criteria reported at the bottom of Tables 3–5. CQQQ, representing the technology sector, exhibits the highest beta, while FXI, which includes the largest 50 companies, has the lowest beta

Source(s): The authors’ estimation

or Create an Account

Close subscription notice
Close access options