Page 46 - ISC PROCEEDINGS 21.4

4. Results and discussion
4.1. Descriptive statistics
Table 2. Descriptive statistics of the variables used
LGDP LFDI LLB LINN
Mean 3.859673 9.749882 1.864101 2.394718
Median 3.861817 9.903090 1.862376 2.511883
Maximum 4.158821 10.30471 1.881248 3.143639
Minimum 3.554372 9.113275 1.844353 1.397940
Std. Dev. 0.183028 0.437023 0.011593 0.557656
Skewness -0.043590 -0.324775 -0.020496 -0.431152
Kurtosis 1.800705 1.454255 1.574252 1.859870

Jarque-Bera 1.747141 3.396918 2.458279 2.469187
Probability 0.417458 0.182965 0.292544 0.290953

Sum 111.9305 282.7466 54.05894 69.44683
Sum Sq. Dev. 0.937983 5.347686 0.003763 8.707442

Observations 29 29 29 29
Source: Research results from Eview 10
Descriptive statistics of the variables are presented in Table 2. The Skewness result
of the variables is negative indicating that the variables are distributed left. At the same
time, Jarque-Bera test indicates that the variables are accepted as normal distributions
4.2. Test statistic
Table 3. Results of PP unit root test
Lag length 0 Lag length 1
Lag length Result
t-statistics p – value t-statistics p – value

LGDP -2.549431 0.3041 -4.726562 0.0008 I(1)
LFDI -2.382916 0.3797 -3.393425 0.0202 I(1)

LLB -1.662213 0.4387 -5.592218 0.0001 I(1)

LINN -1.717052 0.7167 -5.121447 0.0003 I(1)
Source: Research results from Eview 10
To ensure that variables are accepted in the VECM model, it is necessary to test the
stops of the variables. The study performed a stop test using Phillips-Perron (PP). The test
results show that the variables all stop at level 1. That said, the variables that ensure the
VECM analysis is performed
4.3. Cointegration test
To examine the long-term relationship between independent and dependent
variables, the study applied the Johansen co-integrated method.
Table 4. Results of unrestricted cointegration rank test

Unrestricted Cointegration Rank Test (Trace)
Hypothesized Trace 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

45