tag:blogger.com,1999:blog-2198942534740642384.post9084436687935230882..comments2023-10-24T03:16:41.009-07:00Comments on Econometrics Beat: Dave Giles' Blog: VECMs, IRFs & gretlDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger33125tag:blogger.com,1999:blog-2198942534740642384.post-16237877565901478232019-09-07T04:03:53.217-07:002019-09-07T04:03:53.217-07:00Thank you!Thank you!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-88181639428796202352019-09-06T10:36:21.221-07:002019-09-06T10:36:21.221-07:00Amir - I have no idea why they would differ. Perha...Amir - I have no idea why they would differ. Perhaps you should check with the producers of the 2 packages. Ideally, confidence intervals should be given. You could bootstrap them.Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-50448746859854367952019-09-06T09:58:47.035-07:002019-09-06T09:58:47.035-07:00Prof Giles,
Thank you for introducing me to Gretl!...Prof Giles,<br />Thank you for introducing me to Gretl!<br /><br />I have a question regarding IRF for VECM on Eviews and Gretl.<br />On Eviews the IRF (for two variables) gives a decline in Y before a later increase in response to shocks to X. However in Gretl, Y increases in response to shocks to X.<br />I tried my best to apply the same setting regarding VECM. In your opinion, what's the reason for contradictory results?<br /><br />One more thing, Do you believe it is correct to report VECM IRF results from Eviews (even though they don't show confidence intervals) in a paper?<br /><br />I would appreciate your comment!<br /><br />Amir,Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-57947343075221944872014-11-15T02:51:08.251-08:002014-11-15T02:51:08.251-08:00Dear professor Dave,
How do I calculate a...Dear professor Dave,<br /> How do I calculate a simple hedge ratio for futures contracts for hedging a stock portfolio using VECM coefficients.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-71314928483177158962012-11-20T20:06:01.554-08:002012-11-20T20:06:01.554-08:00thank you! have a nice day!thank you! have a nice day!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-21929723612814564222012-11-20T19:56:46.869-08:002012-11-20T19:56:46.869-08:00EViews should be fine.
DGEViews should be fine.<br /><br />DGDave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-87461008984809592612012-11-20T19:53:28.705-08:002012-11-20T19:53:28.705-08:00My final question is: Do you think I can bootstrap...My final question is: Do you think I can bootstrap the CI's using Eviews? Or which statistical tool is easiest for this procedure? Thank you.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-42269760214101315082012-11-20T19:50:46.387-08:002012-11-20T19:50:46.387-08:00Then I guess I have to follow the procedure in thi...Then I guess I have to follow the procedure in this article: A. Benkwitz & H. Lutkepohl, "Comparison of Bootstrap Confidence Intervals for Impulse Responses of German Monetary Systems", Macroeconomic Dynamics, 2001, 5, 81-100.<br /><br />Sigh... :(<br /><br />Thank you anyway :)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-12226372523076411402012-11-20T19:21:01.395-08:002012-11-20T19:21:01.395-08:00Thanks - now I understand! I don't think that ...Thanks - now I understand! I don't think that any of the packages accomodate this. The only option I can see is to bootstrap the CI's.<br /><br />DGDave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-84596730360985329502012-11-20T18:16:09.873-08:002012-11-20T18:16:09.873-08:00Dear Prof.,
Thanks for your reply. IRF using trad...Dear Prof.,<br /><br />Thanks for your reply. IRF using traditional approach is sensitive to the ordering of the variables in the system while generalized approach does not have this shortcoming. <br /><br />Pesaran, M.H., Shin, Y. 1998. Generalized impulse response analysis in linear multivariate models, Economics Letters 58, 17-29.<br /><br />To the best of my knowledge, currently there are only Eviews 6,7 and Microfit do estimate generalized IRF but they do not supply standard errors for VECMs as you say.<br /><br />So my question is if we can calculate Confidence Intervals if I estimate generalized IRFs based on VECM?<br /><br />Thank you very much.<br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-48541499421524138862012-11-20T09:45:55.275-08:002012-11-20T09:45:55.275-08:00Not sure that I understand your question. This pos...Not sure that I understand your question. This post and the previous one deal with C.I.'s for IRF's from VECM models. What do you mean by "generalized" IRF's?Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-66193904022700128852012-11-19T23:39:19.060-08:002012-11-19T23:39:19.060-08:00Dear Prof. Dave,
Thanks a lot for this post. But ...Dear Prof. Dave,<br /><br />Thanks a lot for this post. But may I ask you if it is possible to use this gretl or JMulTi (in another post of yours) to calculate Confidence Intervals if I estimate generalized IRFs based on VECM? <br /><br />Thank you very much. I look forward to hearing from you soon.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-5081853111560169552012-11-04T12:01:43.156-08:002012-11-04T12:01:43.156-08:00Tim - you can, but I'd be somewhat cautious.
D...Tim - you can, but I'd be somewhat cautious.<br />DGDave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-56541573441967299732012-11-04T04:10:46.877-08:002012-11-04T04:10:46.877-08:00Dear Prof. Giles
I`ve got a quick one on the VEC ...Dear Prof. Giles<br /><br />I`ve got a quick one on the VEC model. In case the t- values for the adjustment vector of a VEC model are not statistically significant, can I still use the VEC model and interpret the impulse responses?<br /><br />Best wishes<br />Tim<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-29351636411814130202012-10-16T11:48:23.159-07:002012-10-16T11:48:23.159-07:00I'm afraid I'm not sure. Why not contact E...I'm afraid I'm not sure. Why not contact EViews?Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-77633202857061083242012-10-16T11:46:53.499-07:002012-10-16T11:46:53.499-07:00Not really.....sorry!Not really.....sorry!Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-11788864966397647902012-09-13T10:56:17.468-07:002012-09-13T10:56:17.468-07:00Soryy - don't know off hand, but contact them ...Soryy - don't know off hand, but contact them directly.Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-9415352085301967782012-09-13T10:51:31.213-07:002012-09-13T10:51:31.213-07:00Dear Professor Giles,
thanks for your help, the sy...Dear Professor Giles,<br />thanks for your help, the syntax works in eViews,<br />but I need it for Gretl.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-11461719647283683342012-09-11T09:08:55.242-07:002012-09-11T09:08:55.242-07:00You can get this from the "HELP" (or fro...You can get this from the "HELP" (or from the manual):<br /><br />Syntax <br />var_name.impulse(n, options) ser1 [ser2 ser3 ...] [@ shock_series [@ ordering_series]] <br /><br />Options <br /><br />g (default) <br />Display combined graphs, with impulse responses of one variable to all shocks shown in one graph. If you choose this option, standard error bands will not be displayed. <br /> <br />m <br />Display multiple graphs, with impulse response to each shock shown in separate graphs. <br /> <br />se=arg <br />Standard error calculations: "se=a" (analytic), "se=mc" (Monte Carlo). <br /><br />If selecting Monte Carlo, you must specify the number of replications with the "rep=" option. <br /><br />Note the following: <br /><br />(1) Analytic standard errors are currently not available for (a) VECs and (b) structural decompositions identified by long-run restrictions. The "se=a" option will be ignored for these cases. <br /><br />(2) Monte Carlo standard errors are currently not available for (a) VECs and (b) structural decompositions. The "se=mc" option will be ignored for these cases. <br /> <br />rep=integer <br /> Number of Monte Carlo replications to be used in computing the standard errors. Must be used with the "se=mc" option. <br /> <br />Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-24887601685970810622012-09-11T05:30:42.282-07:002012-09-11T05:30:42.282-07:00Dear Professor Giles,
the syntax to get the impuls...Dear Professor Giles,<br />the syntax to get the impulse- response function of a VECM- model is:<br /><br />vecm order rank ylist ... -- impulse-response<br /><br />It prints the impulse response<br /><br />But how is the syntax (not the click-sequence) to become above graph including the bootstrapped confidence interval<br />Thanks for your helpAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-66595368882057081852012-09-10T09:25:50.301-07:002012-09-10T09:25:50.301-07:00Dear Professor Giles,
the syntax to get the impuls...Dear Professor Giles,<br />the syntax to get the impulse- response function of a VECM- model is:<br /><br />vecm order rank ylist ... -- impulse-response<br /><br />It prints the impulse response<br /><br />But how is the syntax (not the click-sequence) to become above graph including the bootstrapped confidence interval<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-68790827967585833072012-09-10T05:15:40.091-07:002012-09-10T05:15:40.091-07:00Dear Professor Giles,
the syntax to get the impuls...Dear Professor Giles,<br />the syntax to get the impulse- response function of a VECM- model is:<br /><br />vecm order rank ylist ... -- impulse-response<br /><br />It prints the impulse response<br /><br />But how is the syntax (not the click-sequence) to become above graph including the bootstrapped confidence intervalAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-28171405501363979032012-08-26T10:38:50.895-07:002012-08-26T10:38:50.895-07:00Not sure that I understand your question - sorry!Not sure that I understand your question - sorry!Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-80179191891965484982012-08-26T09:57:41.466-07:002012-08-26T09:57:41.466-07:00Dear Professor Giles,
How is the syntax of the im...Dear Professor Giles,<br /><br />How is the syntax of the impulse response from above<br />graph including the bootstrapped confidence interval?<br /><br />Thanks for your help Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-48096923490372537302012-06-21T05:29:44.789-07:002012-06-21T05:29:44.789-07:00Dear Professor Giles,
How we can interpret the im...Dear Professor Giles,<br /><br />How we can interpret the impulse response from above graph?<br /><br />Regards,<br />SASIAnonymousnoreply@blogger.com