Skip to main content

Table 8 Asymmetric BEKK-GARCH model estimation results

From: Half-day trading and spillovers

  \( {R}_{m,t}^c \) \( {R}_{a,t}^c \) \( {R}_{d,t}^c \)
Mean equation
γi, 11 0.013 (0.655) − 0.161*** (− 8.236) 2.487 × 10− 3 (− 0.122)
γi, 12 0.287*** (14.047) −0.038*** (− 2.855) 0.247*** (8.962)
γi, 21 −6.973 × 10− 3 (− 0.466) 2.683 × 10− 3 (0.167) −3.091 × 10− 3 (− 0.354)
γi, 22 −0.049** (− 2.110) −0.045** (− 2.145) −0.050** (− 2.283)
Variance equation
ai, 11 0.170*** (8.341) 0.266*** (18.900) 0.228*** (12.724)
ai, 12 −0.075** (− 2.735) − 0.024 (− 1.252) −0.044*** (− 3.003)
ai, 21 0.025 (1.155) 0.015 (1.440) −0.050 (− 1.490)
ai, 22 0.053 (1.000) − 0.025 (− 0.587) 0.019 (0.440)
bi, 11 0.978*** (232.517) 0.962*** (269.475) 0.972*** (271.051)
bi, 12 0.014** (2.240) 3.587 × 10− 3 (0.737) 0.012*** (3.094)
bi, 21 −0.029*** (− 4.055) −0.011** (− 2.072) −0.041*** (− 4.454)
bi, 22 0.892*** (105.880) 0.897*** (127.187) 0.897*** (120.926)
di, 11 −0.117*** (− 2.846) −0.038 (− 0.970) −0.084** (− 2.369)
di, 12 −0.080*** (− 3.167) −0.091*** (− 3.877) −0.078*** (− 4.879)
di, 21 −0.077** (− 3.512) −0.047*** (− 3.134) −0.127*** (− 4.517)
di, 22 −0.506*** (− 19.320) −0.509*** (− 21.449) −0.489*** (− 19.543)
  1. Notes. This table shows the estimation results of the asymmetric BEKK-GARCH model. The maximum likelihood estimation is applied, and the estimation method is Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm. The results are converged within 100 iterations. The sample period spans from January 1, 2010, to March 31, 2020. \( {R}_{m,t}^c \), \( {R}_{a,t}^c \), and \( {R}_{d,t}^c \) denote different \( {R}_{i,t}^c \) in the mean equation of the asymmetric BEKK-GARCH model. The estimations of the constants in the mean and variance equations are omitted. The t-statistics of the coefficients are shown in parenthesis. One, two and three asterisks (*), respectively, indicate that the t-values are significant at the 0.1, 0.05, and 0.01 level