a

LNME t   0   0 LNMEt 1  1 LNDI t 1   2 LNCPI t 1  3 LNEXRt 1   4 LNPOPt 1  5 LNTOt 1    i LNMEt i

i 1
b c d e f
   i LNDI t i    i LNCPI t i   i LNEXRt i   i LNPOPt i   i LNTOt i  t
i 0 i 0 i 0 i 0 i 0

Stage 1 : Testing ADF and PP Unit Root Test

Level ADF Unit Root PP Unit Root KPSS Unit Root

I(0) Intercept Intercept and Intercept Intercept and Intercept Intercept and
Trend Trend Trend
LNME -1.314 -4.177*** -0.939(16)
LNDI -2.576 -2.706
LNCPI -1.825 -3.480*
LNEXR -0.604 -3.241*
LNPOP
LNTO
First ADF Unit Root PP Unit Root KPSS Unit Root
difference Intercept Intercept and Intercept Intercept and Intercept Intercept and
I(1) Trend Trend Trend
LNME
LNDI
LNCPI
LNEXR
LNPOP
LNTO
Note: 1. ***, ** and * are 1%, 5% and 10% of significant levels, respectively. 2. The optimal lag length is
selected automatically using the Schwarz Info Criteria (SIC) for ADF test and the bandwidth had been
selected by using the Newey–West method for PP and KPSS unit root test.

Stage 2 : Detecting the presence of long run cointegration based on F stat.

Model Max Lag Lag order F statistics

LNME = F(LNDI, LNCPI, LNEXR, LNPOP, LNTO) (4,4) (4,0,3,2,2,1) 4.289**
Critical Values for F stat Lower I(0) Upper (1)
10%
5%
1%
Note: 1. # The critical values are based on Pesaran et al. (2001), case III: unrestricted intercept and no
trend. 2. k is a number of variables and it is equivalent to 5. 3. *, **, and *** represent 10%, 5% and 1%
level of significance, respectively.
Stages 3 : Diagnostic Tests

(A) (B) (C) (D)

Serial Correlation Functional Form Normality Heteroscedasticity
[p-value] [p-value] [p-value] [p-value]
1.637 0.028 7.956 1.196
[0.216] [0.866] [0.018] [0.334]

Note. 1. ** represent 5% significant levels. 2. The diagnostic test performed as follows A. Lagrange
multiplier test for residual serial correlation; B. Ramsey’s RESET test using the square of the fitted
values; C. Based on a test of skewness kurtosis of residuals; D. Based on the regression of squared fitted
values. 2.

CUSUM CUSUMSQ

Stages 5 : Short run elasticities and error correction model (based on present lag)

Short run Elasticities Long run Elasticities

Variables Coefficient Variables Coefficient
LNDI LNDI
LNCPI LNCPI
LNREXC LNREXC
LNPOP LNPOP
LNTO LNTO
ECT C
Note: 1. ***, ** and * are 1%, 5% and 10% of significant levels, respectively.

Stages 6 : Toda Yamamoto Granger Non Causality Test

Malaysia LNME LNDI LNCPI LNREXC LNPOP LNTO

LNME
LNDI
LNCPI
LNREXC
LNPOP
LNTO