[Retracted] Wheat Futures Prices Prediction in China: A Hybrid Approach

Table 2

Descriptive statistics.

	Wheat FP	Rice FP	Real interest rate	Russia wheat FP	Origin	High to low price ratio
Panel A: wheat prospects prices and economic variables summary statistics
JB-statistic	139.326	51.998	114.145	58.149	289.486	1247.737
ADF	−2.285	−1.469	−2.325	−2.407	−4.715	−7.430
DF-GLS	−2.288	−1.326	−1.098	−2.248	−3.726	−7.275
Mean	3.128	3.476	−0.488	11.709	0.015	1.009
Min.	3.033	3.310	−8.080	11.210	−0.089	1.000
Max.	3.225	3.702	6.410	12.160	0.186	1.043
Std. dev	0.055	0.093	2.997	0.227	0.041	0.006
Kurtosis	—	—	2.387	2.390	4.419	7.545
Skewness	0.230	0.368	−0.549	−0.412	1.201	1.796
Panel B: economic variables (first difference) and wheat prospective prices statistics summary
JB-statistic	531.033	0.738	36977.900	267.674	—	—
ADF	−48.144	−45.984	−36.392	−47.781	—	—
DF-GLS	−14.502	−45.826	−4.146	−13.644	—	—
Mean	0.046	0.005	0.085	−0.001	—	—
Min.	−5.316	-3.989	4.030	0.091	—	—
Max.	4.166	3.989	-3.300	−0.096	—	—
Std. dev	0.741	2.011	0.445	0.023	—	—
Kurtosis	5.371	2.985	3.069	0.135	—	—
Skewness	0.081	0.041	30.246	5.108	—	—
Observations	951	951	951	951	951	951

Note. FP means futures price; JB-statistic means Jarque–Bera test; ADF means Augmented Dickey Fuller; DF-GLS means Dickey Fuller generalized least square , , and represent significance level at 10%, 5%, and 1% implying that the null hypothesis is rejected. Null hypothesis for Jarque–Bera test is that the data are normally distributed; null hypothesis for ADF and DF-GLS is that the series has a unit root.