Forecasting the volatility of EUA futures with economic policy uncertainty using the GARCH-MIDAS model

Financial Innovation

Table 1 Descriptive statistics of transformed data

Variables	Mean	SD	Skew	Kurt	JB	Q(5)	ADF	ARCH test
R	0.3e−04	0.032	− 0.805	19.418	64.057***	12.477**	− 47.765***	0.282***
\(^{\bigtriangleup }GEPU\)	0.031	0.220	1.094	5.383	61.510***	9.645*	− 4.106 ***	− 0.051**
\(^{\bigtriangleup }EEPU\)	0.030	0.233	1.238	6.578	111.300***	12.120 **	− 4.524 ***	− 0.040

SD, standard deviation; Skew, skewness; Kurt, kurtosis; JB, Jarque-Bera test; Ljung-Box Q test of lagged 5 order (Q(5)); ADF, augmented Dickey-Fuller unit root test. In the case of the ARCH test, the table reports the coefficient of a first-order lag
***, **, *Indicate significance at the 1%, 5%, 10% levels, respectively