tag:blogger.com,1999:blog-2198942534740642384.comments2017-06-25T20:39:52.003-07:00Econometrics Beat: Dave Giles' BlogDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger3745125tag:blogger.com,1999:blog-2198942534740642384.post-20846190443493422062017-06-25T17:16:17.534-07:002017-06-25T17:16:17.534-07:00Thanks Andy - 2 typos. Now fixed. DGThanks Andy - 2 typos. Now fixed. DGDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-90974661871386449502017-06-25T17:13:03.103-07:002017-06-25T17:13:03.103-07:00I could be wrong, but intuitively I think in the f...I could be wrong, but intuitively I think in the first (ii) above it should be regressing e* on E_1*, and therefore it should be e* in the equation for beta* (instead of y*). My intuition comes from the case where X_2 is uncorrelated with y but X_1 is correlated with y. In that case, y* is zero in expectation, but b is nonzero in expectation.andy whttp://www.blogger.com/profile/16722693090673844654noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-83124211036525534392017-06-25T09:45:03.339-07:002017-06-25T09:45:03.339-07:00The coefficient of the ECT should be negative. Als...The coefficient of the ECT should be negative. Also, this term relates to the short-run dynamics, NOT the long-run relationship.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-29361970934361961462017-06-25T08:57:40.482-07:002017-06-25T08:57:40.482-07:00Hi, Prof. Giles
Im confused over the ECT. After I ...Hi, Prof. Giles<br />Im confused over the ECT. After I found cointegration in Johansen test, I proceeded to VECM, but my ECT was insignificant and it was negative. Does this mean that my result is spurious? As I know, the ECT is referring to long run equilibrium, so in my case, it contradicts with the Johansen test. Can I still proceed to VEC Granger causality and further tests?Unknownhttp://www.blogger.com/profile/16623853514959208048noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-33409529065716187102017-06-23T04:33:12.666-07:002017-06-23T04:33:12.666-07:00Error Correction Model (ECM) Panel Data EVIEWS 9
h...Error Correction Model (ECM) Panel Data EVIEWS 9<br />https://www.youtube.com/watch?v=ZgCwrb6kI7w<br />video Introduce the concept of an Error Correction Model (ECM) Panel Data EVIEWS 9.<br />WhatsApp : +6285227746673<br />PIN BB : D04EBECB<br />IG : @olahdatasemarangOlah Data Semaranghttp://www.blogger.com/profile/12411054888952028992noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-49128596620069545412017-06-21T13:35:20.147-07:002017-06-21T13:35:20.147-07:00Sean - thank you very much.
DGSean - thank you very much.<br /><br />DGDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-70617644356522494452017-06-21T11:13:32.145-07:002017-06-21T11:13:32.145-07:00Hi Dr. Giles,
Excellent post and so glad to see y...Hi Dr. Giles, <br />Excellent post and so glad to see your continuing dedication and hard work in maintaining your outstanding blog. Personally it serves as an incredible resource for myself(An Eviews user) and quite sure many others, as well. Again, thank you. Sean Byrne sean Byrnehttp://www.blogger.com/profile/12370892036071267041noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-27420035686957547842017-06-20T03:40:54.136-07:002017-06-20T03:40:54.136-07:00Having "too many" lags is preferable to ...Having "too many" lags is preferable to having "too few". So (in this particular case) I'd go with AIC model.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-35543105777091365962017-06-19T22:11:55.118-07:002017-06-19T22:11:55.118-07:00Good Day Professor.
I estimated an ARDL model of 6...Good Day Professor.<br />I estimated an ARDL model of 6 variables. After several attempts (using different lags ) to find a better estimate, i got a selected ARDL model using AIC as (1,1,0,0,1,2) while using SIC is ARDL (1,0,0,0,1,2).<br />1. Can I still use this model given these lags selection?<br />2. Which among the AIC and SIC is better?<br />ARDL (1,1,0,0,1,2) = AIC<br />ARDL (1,0,0,0,1,2,) = SIC<br />Thank you sir, in anticipation for your kind acknowledgment<br />Best regards.Miftahu Idrishttp://www.blogger.com/profile/14395040691491376696noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-4056192416885206032017-06-12T06:00:13.146-07:002017-06-12T06:00:13.146-07:00Let's be really clear. The T-Y methodology is ...Let's be really clear. The T-Y methodology is for testing for the presence of Granger causality. That's it. Also, you should NOT use an ARDL model with bounds testing if you have any I(2) variables - that's made very clear in the Pesaran et al. papers. Also - If you have 2 or more I(2) variables you should test if they are cointegrated to an I(1) variable, which in turn may be cointegrated with other I(1) variables you are using. To assist you generally when you have I(2) variables, you might find this helpful: https://ideas.repec.org/a/psc/journl/v4y2012i4p215-252.html<br /><br />Finally, make sure that your variables really are I(2). Often, they are really I(1) but appear to be I(2) because the unit root tests are affected by structural breaks.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-37291835895886445132017-06-12T05:52:59.212-07:002017-06-12T05:52:59.212-07:00It will still be O.K., but it's power may be r...It will still be O.K., but it's power may be reduced somewhat.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-25538506389928778562017-06-12T01:24:33.161-07:002017-06-12T01:24:33.161-07:00When heavily overstating the lag length, will the ...When heavily overstating the lag length, will the Wald test have an asymptotic chi-squared distribution though?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-65710517588043068962017-06-12T00:25:42.728-07:002017-06-12T00:25:42.728-07:00Thank you for the detailed procedure prof. However...Thank you for the detailed procedure prof. However, in our econmetrics class, we were told that causality does not necessary imply impact and equally that in the estimation of the impact of some independent variables on a dependent variable, that in cases where some of the variables are integrated of order two, I(2) one should use the regular VECM or ARDL to estimate the magnitude of impact but instead one should use TODA Yamamoto to estimate the direction of causality. My question then is, how can one estimate the magnitude of impact of the independent variables on the dependent variable in a case where one or more of the variables is integrated of order two, I(2)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-10498095545902577962017-06-11T12:11:01.497-07:002017-06-11T12:11:01.497-07:00Yes, that's fine. The main thing is to make su...Yes, that's fine. The main thing is to make sure that you don't UNDER-state the lag length.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-66190335800702050562017-06-11T09:43:20.957-07:002017-06-11T09:43:20.957-07:00Dear Prof. Giles
I wanted to quickly ask you somet...Dear Prof. Giles<br />I wanted to quickly ask you something: Is the Y-T procedure still valid, even if you choose an arbitrary lag length? For example: AIC tells you to take 8 lags, but the errors are highly correlated. So you increase the number of lags to 12. Is the Y-T-procedure (applied on a VAR with 13 lags then) still valid and superior to Granger causality in differences?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-6282688800330891302017-06-10T05:13:39.935-07:002017-06-10T05:13:39.935-07:00thank u for sharing information with usthank u for sharing information with usRCE Roorkeehttp://www.blogger.com/profile/08454714535524505648noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-59322708999753359392017-06-10T05:04:04.407-07:002017-06-10T05:04:04.407-07:00Thank you for your help! :)Thank you for your help! :)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-49273976532977881542017-06-09T14:20:38.512-07:002017-06-09T14:20:38.512-07:00The whole point of ARDL modelling and bounds testi...The whole point of ARDL modelling and bounds testing is that you can have a mixture of I(0) and I(1) variables. I've stated really clearly, several times, in several posts, that you CAN'T have I(2) variables ion the model.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-76117386826261929322017-06-09T12:36:42.108-07:002017-06-09T12:36:42.108-07:00Would you please tell me that to Run ARDL, it is n...Would you please tell me that to Run ARDL, it is necessary that dependent variable is I(1), or it can be I(0) or I(2)? ThanksUnknownhttp://www.blogger.com/profile/10492956043770853453noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-55465314234391023282017-06-09T02:41:49.468-07:002017-06-09T02:41:49.468-07:00thank uthank uRCE Roorkeehttp://www.blogger.com/profile/08454714535524505648noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-10132786604463367502017-06-08T05:59:18.822-07:002017-06-08T05:59:18.822-07:00Correct! (That's OK :-) )Correct! (That's OK :-) )Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-85596710265290056982017-06-08T02:12:06.512-07:002017-06-08T02:12:06.512-07:00If I find Granger causality and the variables are ...If I find Granger causality and the variables are cointegrated, it is the long-run relationship? Okay, I was mistaken there, I'm sorry. So basically: Granger causality without cointegration refers to the short-run eq, Granger causality with cointegration to the long-run eq (+ VECM coefficients show you the short-run eq) and cointegration without Granger causality to the long-run as well. That should be alright now, isn't it?<br />(I'm sorry for all these questions)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-71924078794088650912017-06-08T01:52:08.997-07:002017-06-08T01:52:08.997-07:00Thank you Prof. Dave for sharing this excellent st...Thank you Prof. Dave for sharing this excellent stuff. Santosh Dashhttp://www.blogger.com/profile/02016226999263087762noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-5650795532578182692017-06-07T15:01:32.197-07:002017-06-07T15:01:32.197-07:00Agreed Dave, It is almost as good as yours!!Agreed Dave, It is almost as good as yours!!not trampishttp://nottrampis.blogspot.com.au/noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-25332992421404853742017-06-07T07:58:30.872-07:002017-06-07T07:58:30.872-07:00Dhanusha - I don't work for EViews - I suggest...Dhanusha - I don't work for EViews - I suggest you contact them through their User Forum at http://forums.eviews.com/Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.com