Page 2 of 2

Re: ARDL bound test

Posted: Sun Dec 27, 2015 11:25 am
by ecofin
GOT IT! thanks Mr igor.

Re: ARDL bound test

Posted: Tue Jan 12, 2016 12:54 am
by metrix
Hi
there is no Unrest linear trend in the menu of ARDL. it's been one year why?
1/can you add this case V (Unrest linear trend).
2/can you make easy way to import the cointegration regression output to the ols object (like ARDL estimation), that we can see the R^2 AIC HQ...DW ...etc, or to see it bellow the cointegration regression (inprove ouput cointegration regression).
3/the bouds test show critical value of F statistic for 4 CASES, in Pesaran & al: BOUNDS TESTING APPROACHES TO THE ANALYSIS OF LEVEL RELATIONSHIPS 2001 p 303 there is tree tabales of t-stistics for CASE I, CASE III, and CASE V. can you add this critical t-statistic for the bounds test.
kind regards.

Re: ARDL bound test

Posted: Wed Mar 09, 2016 10:04 am
by massy
how can we do bound test in eviews 9 if first selected model for ardl has a serial correlation. Because for second best model for ardl we should appy it ,like earlier versions of eviews manual or I dont know how to do it in eviews with new approach. So if I select second model or the reason of seral correlation how can apply it and if estimate it like before should ı do wald test or does it have any new way to do it

Re: ARDL bound test

Posted: Wed Mar 09, 2016 10:34 am
by EViews Gareth
You can fix the lags at whatever values you want.

Re: ARDL bound test

Posted: Wed Mar 09, 2016 11:12 am
by massy
Image

lndexp lnexc lnvol lngdp in dynamic part and ı shoose fixed and ı should write fixed regressors for ardl(1.1.0.0)

I am writing fixed regressors list lndexp(-1) lnexc(-1) lnvol lngdp but not working whats wrong