tag:blogger.com,1999:blog-2198942534740642384.comments2014-09-18T15:29:34.137-07:00Econometrics Beat: Dave Giles' BlogDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger2383125tag:blogger.com,1999:blog-2198942534740642384.post-84147844822820624312014-09-16T19:21:03.434-07:002014-09-16T19:21:03.434-07:00Dear Prof! You have helped a lot for the researche...Dear Prof! You have helped a lot for the researchers throughout the world. Hats off to you with my whole-heart. I am anxiously waiting for your next post for ARDL model in more endogenous varaibles. We all are watching your way for the post. May Allah pak bless you healthy life with prosperity! Thanks again Prof. Niaz Hussain Ghumrohttp://www.iba-suk.edu.pknoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-72801654437298912342014-09-15T07:59:29.484-07:002014-09-15T07:59:29.484-07:00You won't find them. The intercept should be r...You won't find them. The intercept should be retained, even if it is insignificant.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-87189093028041177002014-09-15T04:05:57.374-07:002014-09-15T04:05:57.374-07:00Professor Dave, thank you so much for posting such...Professor Dave, thank you so much for posting such a useful material for us to learn. I have a small question though. I am doing ARDL approach to cointegration and I reached to a situation of no intercept and unrestricted trend after removing highly insignificant variables. But I did not find F-table and t-table to compare my statistics with lower and upper bounds to use the bound test, I have very high F-stat (12.11) though. I know it is significant but still I want go in a formal way. Where can I get such tables? Dambar Upretynoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-11762178194017328792014-09-12T07:20:13.940-07:002014-09-12T07:20:13.940-07:00Gretl is a terrific package and I've blogged a...Gretl is a terrific package and I've blogged about it several times. I'd certainly recommend it for cointegration analysis.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-90091414167740553182014-09-12T06:01:28.288-07:002014-09-12T06:01:28.288-07:00Many thanks, Prof. Giles for the helpful comments ...Many thanks, Prof. Giles for the helpful comments and examples! I have one question - currently I am performing a cointegrtion analysis on economic time series. Could you, please, give your opinion about using GRETL software for this purpose. Thank you!Mariya Neychevahttp://www.blogger.com/profile/05206445780555564234noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-85333086671847192802014-09-10T12:18:16.283-07:002014-09-10T12:18:16.283-07:00In this case they can't be cointegrated. If yo...In this case they can't be cointegrated. If you have 2 or more I(2) series, and one or more I(1) series, you can have cointegration. In this case it's possible for the I(2) to cointegrate to an I(1) series, and the latter can then cointegrate with the other I(1) data. See http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2145744Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-3419925524128336992014-09-10T12:15:25.946-07:002014-09-10T12:15:25.946-07:00You've made it clear that any of the series sh...You've made it clear that any of the series shouldn't be I(2) in order to run ARDL.<br />But, what'd be your suggestion if I need to test for cointegration and my Y~I(1) and X~I(2)?<br /><br />Thanks in advance, I really appreciate what you've done with this blog!Pacohttp://www.blogger.com/profile/05926834696643913334noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-75289933881859512262014-09-04T10:53:54.101-07:002014-09-04T10:53:54.101-07:00The document has now been amended to incorporate t...The document has now been amended to incorporate this important piece of information. Thanks again!Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-45487796071036931592014-09-04T06:45:50.810-07:002014-09-04T06:45:50.810-07:00Gareth - thanks for pointing this out - I should h...Gareth - thanks for pointing this out - I should have run this by you first Ill amend the document accordingly!Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-49760407209773169762014-09-03T15:47:20.805-07:002014-09-03T15:47:20.805-07:00Your point on copying in reverse order to EViews i...Your point on copying in reverse order to EViews is valid, although it should be pointed out that if you copy the date column as well as the values themselves and then paste into EViews, EViews will properly sort them for you, without you needing to do so. EViews Garethhttp://www.blogger.com/profile/02265937096525975321noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-44668198035452530402014-09-02T07:25:48.620-07:002014-09-02T07:25:48.620-07:00R2 relates to the sample variance. It;s just a sam...R2 relates to the sample variance. It;s just a sample statistic.Non-stationarity relates to the population process that generates the data.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-42463488089680412382014-09-02T07:18:10.051-07:002014-09-02T07:18:10.051-07:00Is this really true? The R2 is after all the '...Is this really true? The R2 is after all the 'explained' part of the variance of the 'left-hand-side variable’ (in this case the stock price). But, if the time series are nonstationary, then the variance changes over time?! <br /><br />You probably mean you can use the R2 of the (panel) error correction model?<br />Let’s take the research from above as an example. If you compare the forecasts of analysts for companies from the American region and compare these with forecasts for the European region, can you compare both R2 and make conclusions? If so why?<br /><br />I probably misunderstand the value of not having a ‘spurious’ regression, but I really like to understand it.<br />Vincent Keennoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-73853518919002699232014-08-30T06:27:39.837-07:002014-08-30T06:27:39.837-07:00Esti - you don't need to test dummy variables ...Esti - you don't need to test dummy variables for stationarity - all dummy variables are stationary by construction.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-36634434065362876122014-08-30T03:08:50.123-07:002014-08-30T03:08:50.123-07:00I am so thankful to find your great blog here, Pro...I am so thankful to find your great blog here, Prof Gile! I am doing regression using panel data of 172 regency from 2001-2012. Based on some arguments, i did the unit root test (cause my time series dimension is long enough). I found one of my dummy variable wasn't stationer, can i include it in my model? or i should drop it? <br /><br />Would you please recommend me some references--journal or article? Thank you.Esrihttp://www.blogger.com/profile/14519695064161704877noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-76319318525198402192014-08-29T10:44:17.322-07:002014-08-29T10:44:17.322-07:00No - it's not true, and I've answered this...No - it's not true, and I've answered this at least once before. In the original Pesaran papers it's clear that y can be either I(0) or I(1). Indeed, that the whole point of the "bounds" - one bound is when all variables are I(0) and the other when they are all I(1) - and the includes the dependent variable. You could suggest to your colleague that (s)he should read the original literature!Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-38741376018843368402014-08-29T10:39:34.863-07:002014-08-29T10:39:34.863-07:00It's still valid - this is made quite clear in...It's still valid - this is made quite clear in the original papers. That's where the "bounds" come from - either all data I(0), or all data I(1).Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-52080784075986272312014-08-29T10:33:49.273-07:002014-08-29T10:33:49.273-07:00But it can be either I(1) or I(0) - this is clear ...But it can be either I(1) or I(0) - this is clear in the original papers.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-50954769054797481002014-08-29T03:14:12.635-07:002014-08-29T03:14:12.635-07:00Dear Giles,
One of my colleagues claims that, th...Dear Giles, <br /><br />One of my colleagues claims that, the bounds test can be used only when the dependent variable is I(1) and the explanatory variable(s) is I(0).<br /><br />Is it true?<br /><br />Thanks,<br /><br />Paul W.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-71004256500113870802014-08-29T03:10:42.865-07:002014-08-29T03:10:42.865-07:00Dear Dave,
Thanks for the post, the critical valu...Dear Dave,<br /><br />Thanks for the post, the critical values for the small sample size can be computed by using the technique in "Turner, P. (2006). “Response Surfaces for an F-Test for Cointegration”, Applied Economics Letters, 13:479-482." <br /><br />Regards,<br /><br />veli. Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-84787124745714675762014-08-28T07:56:12.762-07:002014-08-28T07:56:12.762-07:00What matters is the value of T, and T=6 is insuffi...What matters is the value of T, and T=6 is insufficient. Questions such as this would be better addressed to the EViews forum, rather than this blog. See http://forums.eviews.com/Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-53230342238945815622014-08-27T21:21:26.525-07:002014-08-27T21:21:26.525-07:00Dear Professor,
Do you know why we cant conduct u...Dear Professor,<br /><br />Do you know why we cant conduct unit root test in Eviews for panel date with N=20, T=6? It shows insufficient number of observation. But our total observation is 20*6=120 correct? Thanks.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-23786130148606016212014-08-25T11:27:55.773-07:002014-08-25T11:27:55.773-07:00No, you wouldn't.No, you wouldn't.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-485853139503808142014-08-25T11:11:15.241-07:002014-08-25T11:11:15.241-07:00Thanks for the post, Sir. Pls,what happens if the ...Thanks for the post, Sir. Pls,what happens if the bounds test result shows evidence for no long-run relationship (ie. No cointegration relationship)? Do we go ahead to fit a restricted ECM? If yes, what's the rational?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-48032633557683566372014-08-24T18:07:07.053-07:002014-08-24T18:07:07.053-07:00Your point???Your point???Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-4654702744257073292014-08-24T15:55:29.511-07:002014-08-24T15:55:29.511-07:00This result precludes stochastic volatility.This result precludes stochastic volatility.jbhttp://www.blogger.com/profile/03286584518913462346noreply@blogger.com