ARDL bounds test can be used to ascertain the long run cointegration among the study variables of a given model particularly when the unit root of the variables showed mixed order of cointegration (I(0) and I(1)

the Eviews 10.0 has provision for testing for ARDL long run and short run dynamics consequent upon the cointegration.

When the variables in your model are integrated at different orders (say some are I(1) and others I(0)), the ARDL model can be adopted.

Testing for long run relationship using ARDL requires the use of Bound Test as proposed by Pesaran, Smith and Shin (2001). I suggest you read up

http://dx.doi.org/10.1002/jae.616

ARDL reports both long-run and short-run coefficients as reflected in its model. The usefulness or otherwise of the long-run coefficients is however dependent on whether there exist a cointegrating relationship among the variables or not. If after doing Bounds testing for cointegration or any other, and none of the equations is found to be cointegrating, then the relationship reduces to only a short-run.

I hope you find this piece useful. Thanks.

A good summary. @ Paul Terhemba

One of the main test in ARDL approach is BOUNDS TEST, after doing it and the result of F-statistic for this test was higher than the upper critical bound at the significant level (1%, 2.5%, 5% or 10%) then the long run relationship is exists in your model.

I hope this answer is helpful.

You have to perform bound test which indicates if you have a long-run or a short-run of the variables, if there is only short-run you perform the ARDL for the short run , if there is co integration you should proceed with VECM or ECM depends on the results that bound test shows Srikanth Potharla