tag:blogger.com,1999:blog-2198942534740642384.comments2016-02-14T08:46:39.229-08:00Econometrics Beat: Dave Giles' BlogDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger3228125tag:blogger.com,1999:blog-2198942534740642384.post-52554786578865739962016-02-09T09:08:02.860-08:002016-02-09T09:08:02.860-08:00No, it's not, and the same applies for any reg...No, it's not, and the same applies for any regression model.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-69973086480309782312016-02-09T08:29:57.657-08:002016-02-09T08:29:57.657-08:00Thanks for the post. I have a question regarding t...Thanks for the post. I have a question regarding the modeling. Ex what if I have 10 variables where some are strongly correlated and I want to assess the long/short run relationship with a variable y. Is it possible to split the model into two ardl? one for 5 variables and one for the other 5 and still obtain relevant results? I saw a similar question on stackexchange (no answer) so I though I would ask you.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-72860628654495732742016-02-08T07:16:27.915-08:002016-02-08T07:16:27.915-08:00Raluca - no, not at all. They can be positive or n...Raluca - no, not at all. They can be positive or negative, depending on the problem and the sample values.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-47042624116414436592016-02-08T07:05:37.936-08:002016-02-08T07:05:37.936-08:00Hello Professor!
Is it something meaningful or wr...Hello Professor!<br /><br />Is it something meaningful or wrong if one gets negative values for the information criteria like AIC, SIC/BIC or HQ?<br /><br />Regards,<br /><br />RalucaAdminhttp://www.blogger.com/profile/09008629911393135175noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-66393866727008874732016-02-03T02:06:48.118-08:002016-02-03T02:06:48.118-08:00Dear Professor,
you educated many people around th...Dear Professor,<br />you educated many people around the globe with your blog. your amazing job is highly appreciated by many economists all over. i was wondering if you are planning on posting non-linear ARDL examples soon. i noticed it is a new trend and many researchers are starting to adopt it. <br />hope to see it soon.<br />have a wonderful time and enjoyable classes.<br />KyleAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-84588195387961010072016-02-02T14:43:22.181-08:002016-02-02T14:43:22.181-08:00There seems to be some confusion here. There is no...There seems to be some confusion here. There is no LM test for multicollinearity. You could test for homoskedasticity. The Q statistics tell you about the serial independence of the errors - nothing to do with heteroskedasticity.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-57496116120290499932016-02-02T09:54:10.045-08:002016-02-02T09:54:10.045-08:00If you meant to say "to the dependent variabl...If you meant to say "to the dependent variable", then the answer is "yes".Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-73079386719713681182016-02-02T09:53:15.416-08:002016-02-02T09:53:15.416-08:00If you have 4 breaks for the same variable, then t...If you have 4 breaks for the same variable, then this implies 5 different "regimes" and you will need to construct 4 separate dummy variables.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-24422596190553987602016-02-02T05:05:02.175-08:002016-02-02T05:05:02.175-08:00As the sum of the standardized coefficients is 1, ...As the sum of the standardized coefficients is 1, can we interpret on the scale of 1, the size of the coefficient is weight (contribution) of that independent variable?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-41660321546933760172016-02-02T02:08:03.364-08:002016-02-02T02:08:03.364-08:00if we have multiple break for different regressor,...if we have multiple break for different regressor, for example 1994, 1999, 2001, 2004 for 4 different regressor, is it enough to create 1 dummy variable that include all the break? Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-36396637248855682972016-02-02T00:17:12.719-08:002016-02-02T00:17:12.719-08:00Thanks for the good info Prof. My questions, is it...Thanks for the good info Prof. My questions, is it necessary to run a LM test for testing and detecting the multicolinearity and heteroskedastcity test before we running the bound test and cointegrating and long run test? or its just enough by looking the Q stats.<br /><br />if there is no long run relationship based on bound test, then what next? are we going to stop there and then give up? <br /><br />Thanks Prof.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-58922300337798422142016-02-01T09:58:27.528-08:002016-02-01T09:58:27.528-08:00No, you can't.No, you can't.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-67916156039505058312016-02-01T09:35:29.169-08:002016-02-01T09:35:29.169-08:00Hello professor Giles
I want to apply SURADF and C...Hello professor Giles<br />I want to apply SURADF and CADF test in Gauss or R. But i couldnt because i didnt use GAUSS or R before. Could you help me about this subject?Özge BARIŞ TÜZEMENhttp://www.blogger.com/profile/16956393743552836790noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-20154204725095668932016-02-01T09:03:59.602-08:002016-02-01T09:03:59.602-08:00Dear Prof. Gelis
Thank you for this very useful po...Dear Prof. Gelis<br />Thank you for this very useful post<br />I would like to ask that if I find R2 > DW, I can directly decide it is a “spurious regression”. I mean it is almost enough.<br />Thank you again<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-59191992321063231592016-01-29T09:44:36.259-08:002016-01-29T09:44:36.259-08:00I think you should check the EViews Forum, at: ht...I think you should check the EViews Forum, at: http://forums.eviews.com/Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-88233026893037311542016-01-29T04:23:47.663-08:002016-01-29T04:23:47.663-08:00Hopefully this will also (eventually?!) become a t...Hopefully this will also (eventually?!) become a trend in developing economies as well. When I enrolled in my PhD studies I knew that there exists something that's called "econometrics". Yet, as the time gone by, I've realised that if you want to have an actual impact in some field of economics, econometrics is "a must have" type of knowledge. So, I'm trying my best to improve my econometric skills every day, unfortunately only on a self-learning basis since a course or basically anything (seminar, workshop, etc) regarding contemporary issues in econometrics is an inconceivable thing in my country. Anyway, I'm not discouraged by that, and I will pursue my self-learning process, and I just want to add that this blog has helped me a lot so far. So, thank you for all your great posts, they are invaluable for students like me.<br />P.S. I've been a regular visitor for quite a while now. Yet, this is my first comment (since I've found answers to all of my previous questions somewhere between the lines :)Jugnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-86491741771084422782016-01-28T21:23:36.475-08:002016-01-28T21:23:36.475-08:00No, I'm using EViews 8.1. Is there also an ava...No, I'm using EViews 8.1. Is there also an available option to use student's t distribution similar to part 2 of your example in EViews 9?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-87321634127981188922016-01-28T20:20:01.041-08:002016-01-28T20:20:01.041-08:00Renie - you don't need to use a LOGL object. I...Renie - you don't need to use a LOGL object. I am assuming you are using EViews 9. Just specify your equation as if you were going to estimate it by OLS. Indeed, choose OLS as the estimator, but in the "options" tab, select "ML" as the estimator.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-55818936233895555292016-01-28T19:36:24.257-08:002016-01-28T19:36:24.257-08:00Dear Professor,
I am trying to estimate an ARMA mo...Dear Professor,<br />I am trying to estimate an ARMA model using MLE. I am having a hard time since the eps series requires its own past values and it won't run. Is it possible to estimate an ARMA model using logl object in eviews? And how could I fix this problem if it is possible?<br />Thank you very much,<br />RenieAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-61815955787466491972016-01-27T12:40:49.312-08:002016-01-27T12:40:49.312-08:00thanks for your quickly answer Mr.Giles, and it is...thanks for your quickly answer Mr.Giles, and it is great that you share your knowledge.jesser-utmachhttp://www.blogger.com/profile/17900677201827676642noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-56486205756339287222016-01-27T12:22:22.177-08:002016-01-27T12:22:22.177-08:00Yes. (If the data are all stationary, they can'...Yes. (If the data are all stationary, they can't be cointegrated.)Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-6898176647658367682016-01-27T12:21:11.837-08:002016-01-27T12:21:11.837-08:00Hello and congratulations for your great and helpf...Hello and congratulations for your great and helpful post. A question: Can I apply a VARM at level with stationary data but with no cointegration? thanks in advance.jesser-utmachhttp://www.blogger.com/profile/17900677201827676642noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-43868148711506510252016-01-27T10:50:07.498-08:002016-01-27T10:50:07.498-08:00Would "The Elements of Statiscal Learning&quo...Would "The Elements of Statiscal Learning" count as an econometrics textbook? Perhaps not, but there are very relevant topics, I believe. It is freely available at the authors' website, e.g. here: http://statweb.stanford.edu/~tibs/ElemStatLearn/. The sister book "An Introduction to Statistical Learning" which is a little more basic is also freely available: http://www-bcf.usc.edu/~gareth/ISL/.Daumantasnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-1865760599614043032016-01-26T18:13:32.417-08:002016-01-26T18:13:32.417-08:00Hi Professor Giles,
I was curious about your thou...Hi Professor Giles,<br /><br />I was curious about your thoughts on this article: https://blogs.cfainstitute.org/investor/2016/01/26/a-valuation-method-for-private-equity/<br /><br />These folks claim to have built a better mouse trap for the valuation of private equity firms. While they don't go into a huge amount of detail regarding their methodology (i.e. the exact specification of their econometric model) they throw around things like "We have really high R^2". To me, at a surface level, it seems like they may be simply overfitting data. With that said, I would be very curious to hear your thoughts on this.<br /><br />Regards,<br />DarrylDarrylnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-24243554183975811702016-01-26T10:36:14.484-08:002016-01-26T10:36:14.484-08:00Fitting the parameters a and b and introducing a t...Fitting the parameters a and b and introducing a time lab of one data set wrt the other seems to show a good correlation between core CPI and CLF via the following relation:<br /><br />log CPI = b + a log CLF <br /><br />Here's some examples:<br /><a href="http://4.bp.blogspot.com/-2Ij2oURmRvs/VqP0x04k4xI/AAAAAAAAIpI/y3vmQlJPuZM/s1600/ite%2Bmodel%2Bfor%2Bcpi%2B%2528us%2529%2B1.png" rel="nofollow">USA</a><br /><a href="http://3.bp.blogspot.com/-ks18NmLHsyk/VqLzYPt3KOI/AAAAAAAAIoY/AT8q_u1_zh8/s1600/ite%2Bmodel%2Bfor%2Bcpi%2B%2528japan%2529%2B2.png" rel="nofollow">Japan</a><br /><a href="http://3.bp.blogspot.com/-ONA7iVOaXx0/VqP3FamAz9I/AAAAAAAAIpg/sgBytSymMiA/s1600/ite%2Bmodel%2Bfor%2Bcpi%2B%2528canada%2529%2B2.png" rel="nofollow">Canada</a><br /><br />In your view, is this a good candidate for a Granger causation analysis?Tom Brownhttp://www.blogger.com/profile/17654184190478330946noreply@blogger.com