Skip to main content
Financial Innovation

Table 2 Stationary and residual diagnostic tests

From: Cue the volatility spillover in the cryptocurrency markets during the COVID-19 pandemic: evidence from DCC-GARCH and wavelet analysis

 

BTC

ETH

XLM

XRP

USDT

ADA

LTC

EOS

ADF (Level)

 − 1.676

2.660

0.791

 − 3.384

 − 3.329

0.605

 − 0.751

 − 2.872

 

(0.760)

(1.000)

(0.999)

(0.055)

(0.063)

(0.999)

(0.967)

(0.173)

ADF (1st difference)

 − 4.692

 − 5.071

 − 6.613

 − 13.87

 − 14.39

 − 4.093

 − 6.551

 − 21.99

 

(0.001)

(0.001)

(0.000)

(0.000)

(0.000)

(0.007)

(0.000)

(0.000)

Q-statistics

7696.2

2949.4

5826.2

5171.9

329.8

6590.0

6785.8

4761.3

 

(0.000)

(0.000)

(0.000)

(0.000)

(0.073)

(0.000)

(0.000)

(0.000)

Normality test

852.8

926.5

1061.4

1246.5

42,805.3

769.5

802.0

520.6

 

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

LM test

1625.4

1208.7

615.5

379.9

3.059

818.8

660.3

160.1

 

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

ARCH effect

1096.1

955.4

285.2

136.5

27.27

751.4

540.1

102.8

 

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

(0.000)

  1. p-values are given in parentheses. For the Augmented Dickey-Fuller (ADF) test, the trend and intercept are included in the test equation. The lag length is selected through the Akaike Information Criterion (AIC). The lag length for the Q-statistics is selected as 36. The normality test statistics reflect the Jarque–Bera test statistics. The null hypothesis in the LM test is no serial correlation at up to 36 lags. The ARCH effect is selected to detect heteroskedasticity among the series, which regresses the squared residuals on lagged squared residuals and a constant at up to 36 lags
Back to article page