Estimation ARDL model in Eviews9

[thomzar]

Hello…I am new to ARDL model and Eviews9.
I am estimating the equation: export = c + GDP + Real exchange rate + Exchange rate volatility
I am using the ARDL model and Eviews9.
May I have two questions?

(Q-1). why is the value of SR coefficient from Bounds Test and from the Long Run estimation different? Should they be the same values?
Based on the attached files: SR coefficient of D(LEXPO(-1)) = -0.136228 (from Bounds Test), but it is -0.113048 (from Cointegration and long run).

(Q-2) Based on question-1 above, which coefficient is considered as SR coefficient? -0.136228 (from Bounds Test)? Or -0.113048 (from Cointegration and long run)?
Thank you very much.

[notime]

the result from the co integrating and long run form gives the SR coefficients in the first part while the LR coefficients are given in the second part of result. Your bounds test reveals that there is no long relationship among the variables

[thomzar]

Hello…thank you very much for your kind advices. I also think that the first part is SR coefficient and my results show no cointegration. However, my question is that, based on my understanding, the result from ARDL Cointegration and Long Run Form is similar to the result from ARDL Bounds Test. The different is that the LR coefficients of ARDL Cointegration and Long Run Form come from the LR coefficient of Bounds Test, but already normalized (divided through) by coefficient of export (LEXPO(-1)). The SR coefficient in ARDL Cointegration and Long Run Form should be equal to the SR coefficient from Bounds Test since we do not do anything with it.

Based on the example in Youtube which is provided by HIS Eviews from minute 12.00 ( https://www.youtube.com/watch?v=542uZVN ... zF&index=5 ) the SR coefficient of ARDL Cointegration and Long Run Form are equal to the SR coefficient from Bounds Test.

I very confuse that why I obtain the different values of SR coefficients. Thank you very much.