Evaluating Co-movements with the Total Value of Import and Export with Exchange Rate in Yunnan Province - Innovative Management in Information and Production - page 43

EXCHANG RATE IMPORT AND EXPORT

.012 .3

.2

.008

.1

.004

.0

.000

-.1

-.004

-.2

-.008 -.3

01 02 03 04 05 06 07 08 09 10 11 01 02 03 04 05 06 07 08 09 10 11

Fig. 2 The incremental rate for the monthly total value of import and export with exchange rate

Tabl e 1 Summary statistics
for the value of imports and
exports and exchange rate

Exchange rate Import and export

Mean (%) 0.000875 0.008085

SD (%) 0.002530 0.091245

Skewness 0.571722 − 0.199908

Excess Kurtosis 3.247830 2.762848

Max (%) 0.009450 0.208290

Min (%) − 0.004960 − 0.223460

JB 7.471814 1.179514

Tabl e 2 Tests of hypotheses of unit-root

ADF PP

Log of first
difference

Log of first
difference Variables Level Level

5.851011 ∗∗ 5.857011 ∗∗

Exchange rate − 1.308496 − − 1.308496 −

14.37424 ∗∗
Note: The ADF, the log of first difference ADF, PP, and the log of first difference PP tests should
be compared. The critical values for the rejection of the null hypothesis of a unit-root are

5.466175 ∗∗ 14.37424 ∗∗ 5.466175 ∗∗ Import and export − − − −

− 4.0295

and 3.444 for 1% and 5%, respectively. The log of first difference series is at the 1% significance

level. The symbols ∗∗ and ∗ denote rejection of the null hypothesis at the 1% and 5% significance
levels, respectively.

−

The total value of import and export with exchange rate should be stationary for
modeling the time series, thus testing for unit roots is essential for the time series
analysis with multivariate GARCH models. ADF and PP test can perform the test
for unit-root; Table 2 shows the results of strong support of the null hypothesis 8 of
unit-root for the first difference of log-transformed.
Tab le 3 reports the maximum likelihood estimates of the GARCH model. The
ARMA (1.0)-GARCH (1.1) specification is the best model for exchange rate, while
export and import are appropriate of ARMA (1.1)-GARCH (1.1) All autoregressive
coefficients

are highly significant. All the positive values of x 1 imply that it is
positive related to the previous one. The result of the conditional variance equations
is ˆ

β

ˆ

993, 0.999 for the monthly total value of import and export with
exchange rates, respectively. As can be seen in the variance equation, the asymmetry
parameters,

α + β = 0 .

λ i , are significant and positive and skewed to the right.