When to use OLS and ARDL?

[noor ana]

Hi there. I am very very new to STATA and in fact new to data analysis techniques. But I have attended a course on STATA and now in the phase of choosing the right technique for my data. But I am a little confused with the long run and short run data. That should tell me to choose between using OLS or the ARDL. My data consists of 16 banks in 6 years time. So far I have not run the data in STATA since there are still some problem in the dataset itself. But I will appreciate it alot if anyone here would like to explain on the difference? Maybe I should ask what is the difference between regression on level and regression on lag? Is it? Tq in advance.

[E. David Aja]

The following post (by Jeff Wooldridge) should help illuminate some of the confusion:

http://www.statalist.org/forums/foru...864#post517864

In short, ARDL is a kind of model which can be estimated via OLS (though it may not be the best method to do so).