tag:blogger.com,1999:blog-2198942534740642384.comments2017-02-22T10:44:55.554-08:00Econometrics Beat: Dave Giles' BlogDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger3604125tag:blogger.com,1999:blog-2198942534740642384.post-90858206252971603542017-02-21T16:38:22.681-08:002017-02-21T16:38:22.681-08:00Thanks - that was quick :) Thanks - that was quick :) Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-31638410699472906802017-02-21T16:29:43.217-08:002017-02-21T16:29:43.217-08:00Thanks. Yes, I think that's how I'd interp...Thanks. Yes, I think that's how I'd interpret that result.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-46410783542552246132017-02-21T16:22:28.916-08:002017-02-21T16:22:28.916-08:00Hi Dave – You’re doing awesome work
Question:
I ra...Hi Dave – You’re doing awesome work<br />Question:<br />I ran the cointegrating and Long run model and obtained an error-correction coefficient of -0.5614 which was negative and significant.<br />The long run coefficients were not significant. What is the meaning/interpretation of this?<br />Does it mean that the two series just move together, but one cannot explain the other? <br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-12940417470345721232017-02-17T09:16:19.467-08:002017-02-17T09:16:19.467-08:00Thank SirThank SirAbdur Rehmanhttp://www.blogger.com/profile/11040447266519309166noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-13739396560248756832017-02-11T07:11:37.783-08:002017-02-11T07:11:37.783-08:00I understand your concern - in this case, take a l...I understand your concern - in this case, take a look at the editorial board and the authorship of the papers published there recently.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-24305994616828263642017-02-09T06:33:51.991-08:002017-02-09T06:33:51.991-08:00The journal Econometrics may be the exception, but...The journal Econometrics may be the exception, but overall, MDPI journals are of dubious quality and I would recommend sending your work there. There are many, many good econometrics journals out there. Be skeptical of these new open access journals.<br />http://www.universityaffairs.ca/features/feature-article/beware-academics-getting-reeled-scam-journals/Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-66215488045212537252017-02-05T13:39:15.777-08:002017-02-05T13:39:15.777-08:00You need to set a maximum lag length for every var...You need to set a maximum lag length for every variable. Then you consider every possible combination of lag lengths for the various variables, and choose the combination that minimizes SIC (or AIC if you prefer). This is likely to be a lot of combinations, so write a few lines of program code to do it.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-81168064257164065962017-02-05T13:16:35.080-08:002017-02-05T13:16:35.080-08:00Dear Professor, Thanks a ton for the superb articl...Dear Professor, Thanks a ton for the superb article. One thing that I am having great difficulty with is that how to select the optimum number of lags for the dependent and independent variables. In your article you have used just one independent variable to find out the lag length. Can you please elaborate it a bit in case of using Eviews 8.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-41439998405534213842017-01-31T10:53:48.606-08:002017-01-31T10:53:48.606-08:00Dear Professor,
First of all, I would thank you f...Dear Professor,<br /><br />First of all, I would thank you for a really nice job on your blog Econometrics Beat. It´s amazing what you do there. <br /><br />Because of this, i took a liberty to ask you something about impulse response analysis. I have doubts about the interpretation of this plots and i´m quite sure that is a common doubt. <br /><br />Despite of that, i couldn´t find any exact answer for this question. The problem is what the criteria for analysis the statistical significance of a response in a variable with a shock impulse? <br /><br />What I mean, if I have plots like on the figures in attachment, how can I know if there is statistical significance? The lower and upper confidence bands are both positive or negative is a criteria to statistical significance?<br /><br />For example, in your blog you put a figure with IRF plot (http://davegiles.blogspot.com.br/2013/04/confidence-intervals-for-impulse.html) where lower band is negative and upper band is positive. Is this a criteria for not significantly, like you said in comments?<br /><br />If you could stablish a criteria for this type of analysis or indicate a paper/slides/post where this kind of doubt is treated I´ll be very, very, helpfull. <br /><br />Thank you for attention!<br /><br />https://www.dropbox.com/s/chq0sp42m1xhwev/plot1.png?dl=0<br /><br />https://www.dropbox.com/s/qwibyuru906gxe6/plot2.png?dl=0Rodrigo Loureirohttp://www.blogger.com/profile/01256229284384524999noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-66528152852108221652017-01-30T10:48:20.883-08:002017-01-30T10:48:20.883-08:00Dear Prof.Dave,
Thank you for your helpful blog, I...Dear Prof.Dave,<br />Thank you for your helpful blog, I am trying to estimate an ARDL model relative to 2 time series data with Eviews-9.5. I've two questions: <br />1- Is it right to look for structural changes in the ratio of the two variable with Bai and Perron approach, then we introduce Dummy variables on the identified break dates in the ARDL model.<br />2- If yes, how we construct dummy variables that deal with multiple change points; for example if break dates in the ratio are in 1997, 2001 and 2014? is it correct to consider one dummy variable that takes 1 in the specified dates (1997, 2001 and 2014) and 0 outside?<br />3- If the answer of question 1 is NO, can I apply the Breakpoint URT test in eviews for the search of multiple break points in order to introduce them in the ARDL? and how to define the dummy variable in the case of different break dates (example in X, 2000 and in Y, 2010) in the two series?<br />thank you<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-90737538043019185792017-01-26T13:09:10.866-08:002017-01-26T13:09:10.866-08:00Maria - if you have monthly data then HEGY is not ...Maria - if you have monthly data then HEGY is not appropriate. The appropriate testing is described by Beaulieu and Miron, J. Econometrics 1993. See this link:<br />http://www.sciencedirect.com/science/article/pii/030440769390018ZDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-83262370939290221352017-01-26T11:13:19.494-08:002017-01-26T11:13:19.494-08:00I read that HEGY was written to deal with quarterl...I read that HEGY was written to deal with quarterly data. What are the implications of running this test on monthly data??? María Mercedes Vanegas Cantarerohttp://www.blogger.com/profile/13454459368650423831noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-52161407294471260472017-01-24T08:38:02.284-08:002017-01-24T08:38:02.284-08:00Maria - this old post may be of some help. You sho...Maria - this old post may be of some help. You should also check the literature on testing for nonlinear cointegration. For example, see http://yoda.eco.auckland.ac.nz/nzesg/PDFs/paper/Haug.pdf<br />and also https://ideas.repec.org/a/bes/jnlbes/v19y2001i3p331-40.htmlDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-24496091901583724092017-01-24T03:09:08.868-08:002017-01-24T03:09:08.868-08:00Dear Prof. Giles,
Could you guide me to some liter...Dear Prof. Giles,<br />Could you guide me to some literature on how to test Granger causality on nonlinear models? I am a bit skeptical on whether my results are valid or not currently (perhaps a linear model is not the best fit). I have a stable VAR, no autocorrelation present and both variables are I(1). Ultimately, I found no cointegration between the variables and no causality. I wonder if this could be because of the linear assumption. The problem is, I don't know how to test Granger causality on nonlinear models and can't seem to find good literature on the subject. Thank you in advance!María Mercedes Vanegas Cantarerohttp://www.blogger.com/profile/13454459368650423831noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-81322856609285067382017-01-21T23:08:56.992-08:002017-01-21T23:08:56.992-08:00Thank you very much Dear Professor!Thank you very much Dear Professor!Faisal Sherhttp://www.blogger.com/profile/16355528359488434880noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-65739435126227487032017-01-21T12:34:03.542-08:002017-01-21T12:34:03.542-08:00Faisal no, you can't do this with an ARDL mod...Faisal no, you can't do this with an ARDL model. You need to use a VAR model with your I(1) variable first-differences, and the other variables in levels. (You can also difference the I(0) variables if you wish.)Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-52931036215689039272017-01-21T12:12:53.047-08:002017-01-21T12:12:53.047-08:00This comment has been removed by the author.Faisal Sherhttp://www.blogger.com/profile/16355528359488434880noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-47489379674221413072017-01-19T16:12:45.273-08:002017-01-19T16:12:45.273-08:00This comment has been removed by the author.Adam Elderfieldhttp://www.blogger.com/profile/14269270642659046827noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-64425741494809548652017-01-19T00:36:51.511-08:002017-01-19T00:36:51.511-08:00Thank you Professor Giles for sharing this valuabl...Thank you Professor Giles for sharing this valuable book with your students. It's really a great book. It will help practitioners a lot.Santosh Dashhttp://www.blogger.com/profile/02016226999263087762noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-53049351523568704932017-01-18T08:11:43.164-08:002017-01-18T08:11:43.164-08:00Again - bootstrap them. I'll see if I can do b...Again - bootstrap them. I'll see if I can do better than this, though.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-77265942944353342862017-01-18T07:42:10.721-08:002017-01-18T07:42:10.721-08:00Hi Dave! I was wondering if there is a way to calc...Hi Dave! I was wondering if there is a way to calculate prediction intervals around predicted values from orthogonal regression? Thanks Ramnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-18299667532422840952017-01-17T14:30:38.653-08:002017-01-17T14:30:38.653-08:00Neither of the two cases should work. In both cas...Neither of the two cases should work. In both cases (given the same regression model), the intercepts would be (α + y), (α - y) in the first case & (α), (α +y), (α - y) in the second case. Both of these are assuming a negative effect of the lack of condition A or condition C respectively.<br /><br />The first case has a whiff of plausibility however. If there existed a condition such that effects were inherently positive when it holds and inherently negative when it does not (or vice versa), perhaps it would work. An example may be the existence of debt and its effects on one's credit score. It does still seem to be making assumptions before testing the data...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-77551344748764328172017-01-14T14:08:21.883-08:002017-01-14T14:08:21.883-08:00Julio - yes, they certainly are.Julio - yes, they certainly are.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-32879773964007196292017-01-14T13:37:05.250-08:002017-01-14T13:37:05.250-08:00Thank you very much for your explanation. So, ther...Thank you very much for your explanation. So, there is no option when there are not observations after period T. Dynamic forecasting is used, however error prediction is higher. My question is: what about indicators such as (i) root mean squared error, (ii) mean absolute percent error, (iii) bias proportion, (iv) variance proportion. Are they helpful in order to know the forecast performance?Julio Rospigliosihttp://www.blogger.com/profile/15485918486781230399noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-74928504044446540292017-01-12T14:44:42.530-08:002017-01-12T14:44:42.530-08:00If you use the forecast function, clicking static,...If you use the forecast function, clicking static, and choose to forecast the variable in levels rather than differences this will "unscramble" it. Alternatively, you could use the output and work through the maths in a spreadsheet, which will help remove the "black box" illusion of eviews. Adam Elderfieldhttp://www.blogger.com/profile/14269270642659046827noreply@blogger.com