Skip to main content
Financial Innovation

Table 3 Parameters of selected DCC-GARCH models

From: The time-varying causal relationship between the Bitcoin market and internet attention

Variable

(eGSVI(t), e(t − 1))

(eAGSVI(t), e(t − 1))

(eGSVI(t), eVolume(t − 1))

(eAGSVI(t), eVolume(t − 1))

a1

0.009***

(0.001)

0.007***

(0.001)

0.009***

(0.002)

0.008***

(0.002)

a2

0.523***

(0.076)

0.990***

(0.065)

0.061***

(0.076)

0.078***

(0.069)

A11

0.319***

(0.092)

0.246***

(0.090)

0.2990***

(0.097)

0.272***

(0.099)

A22

0.132**

(0.207)

0.157**

(0.320)

0.108***

(0.021)

0.120***

(0.024)

B11

0.260***

(0.028)

0.435***

(0.034)

0.273***

(0.029)

0.328***

(0.030)

B22

0.846***

(0.031)

0.789***

(0.045)

0.516***

(0.146)

0.404***

(0.165)

dcc.alpha

0.027***

(0.014)

0.022***

(0.015)

0.0018*

(0.006)

0.000

(0.005)

dcc.beta

0.908***

(0.055)

0.904***

(0.086)

0.980***

(0.098)

0.987***

(0.393)

  1. ***, **, * denote statistical significance at a 1%, 5%, and 10% level, respectively. e(t), eGSVI (t), eAGSVI (t) and eVolume(t) are the innovations of Bitcoin returns, dlog_GSVI, AGSVI and dlog_volume at time t, which are estimated from ARMA-GARCH models. (eGSVI(t),e(t − 1)) represents DCC-GARCH model of eGSVI(t) and e(t − 1)
Back to article page