tag:blogger.com,1999:blog-2198942534740642384.comments2016-10-24T01:32:54.403-07:00Econometrics Beat: Dave Giles' BlogDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger3481125tag:blogger.com,1999:blog-2198942534740642384.post-89639633248171722612016-10-21T12:52:50.885-07:002016-10-21T12:52:50.885-07:00The same as for any regression model - it's th...The same as for any regression model - it's the number of observations minus the number of regressors.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-58890065526495028262016-10-21T02:57:57.969-07:002016-10-21T02:57:57.969-07:00Please proof , how to check the degree of freedom ...Please proof , how to check the degree of freedom in ARDL MODEL?amkhttp://www.blogger.com/profile/15867500960790926409noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-8985757097866213302016-10-20T07:10:00.466-07:002016-10-20T07:10:00.466-07:00Camilla - there is no book!! The picture was a spo...Camilla - there is no book!! The picture was a spoof. I thought that everyone realized that!!!!! Responses to your other questions will follow. DGDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-82867151785043896032016-10-20T07:05:24.286-07:002016-10-20T07:05:24.286-07:00Dear Dave Giles, I have two (or three) questions r...Dear Dave Giles, I have two (or three) questions related with the dummy for dummies - First is - where can we get a copy of the book by 2016 (since I cannot find it on Amazon)? <br /><br />Second question, I often use models with a lot of dummy variables for time, industry, country, ownership etc. Is there an easy way to interpret these coefficients - for example should I use individual or combined dummies for countries and time in order to interpret something about the effect of the financial crisis on exports before and after (I have the years 2007 and 2011 and I currently work with a small sample of four countries). Also we would ideally like to say something about the relationship between exchange rate system and exporting at the firm level. But I read the paper by Wooldridge (2005) on the problems of statistical inference for institution or country level dummies (I can send you the full refernece if you need it). So I suppose that even though we have 5000+ observations at the firm-level we cannot say anything in terms of statistical significance/certainty - e.g. that firms in countries with fixed exchange rates fare better in the long run etc. I think this is more like a case study at the institutional/country level and there is no reason to insert anything but country dummies (e.g. not changes in currencies over time), since there is no point after all as the interpretation of those country effects would have no statistical meaning after all? <br /><br />Hope you can help us and also if you have an easy guide (maybe it is in the book) when we combine different dummies and it seems very difficult to interpret them when I read Wooldrige's introductory textbook. <br /><br />Also, I had one last question related with your original post from 2011 - do you think it ok in those semi-log equations just to comment on the significance of the dummy variables. I think often we are not so concerned about their absolute size or impact as it can be difficult to interpret consistently in the context of the model or would you as a reviewer of a paper consider that very sloppy? <br /><br />Thank you so much for your patience! <br /><br />CamillaCamilla Jensenhttp://www.blogger.com/profile/15581667516870037624noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-5002785517708805612016-10-18T06:15:04.548-07:002016-10-18T06:15:04.548-07:00Why not check out the STAT forum - e.g., here: htt...Why not check out the STAT forum - e.g., here: http://www.statalist.org/forums/forum/general-stata-discussion/general/1355211-ardl-panel-model-in-stataDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-45137082245140011262016-10-18T05:56:42.548-07:002016-10-18T05:56:42.548-07:00No you can't. The paper you mentioned (and lot...No you can't. The paper you mentioned (and lots of others out there) are just plain wrong. Applying the T-Y approach is dead simple.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-25604810070089087932016-10-18T02:28:49.635-07:002016-10-18T02:28:49.635-07:00Hi Dave,
I am using a series on number of Europe...Hi Dave, <br /><br />I am using a series on number of European union directives as an independent variable over a long historical time series. The series is zero until the 1970s when countries join the European union. The zeros are genuine and not missing values but do reflect a change in regime. I’m not sure how best to handle – I have tried a dummy for when countries entered the EU but am missing out on the growth after the 1970s. I’ve seen a lot of ways suggested to handle having lots of zero in a time series but I'm not sure what the best one is. <br />Grateful for any thoughts,Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-62212564045779793102016-10-18T02:14:49.961-07:002016-10-18T02:14:49.961-07:00Dear Prof. Giles,
Thank you for this very informat...Dear Prof. Giles,<br />Thank you for this very informative post. <br />I'm interested in using a bound testing approach for a panel data series. Can I still use this approach? <br />If so, how would I test for the optimal lag length, serial correlation and stability of the model? I'm using the Stata software.<br />Thank you very much<br />AshaniAshanihttp://www.blogger.com/profile/15823119418071014562noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-27154545918379645452016-10-17T15:15:26.994-07:002016-10-17T15:15:26.994-07:00Can I just run a Granger test between two I(1) ser...Can I just run a Granger test between two I(1) series in a VAR environment? I only need to see the lead and lag relationship between the two. <br /><br />Basically that's what this paper did "The crude oil market and the gold market: Evidence for cointegration, causality and price discovery"Jamiehttp://www.blogger.com/profile/15372649750664976487noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-79755791032477288992016-10-17T06:06:23.616-07:002016-10-17T06:06:23.616-07:00Only if you have lots of observations. It;s just a...Only if you have lots of observations. It;s just a regression model you need positive degrees of freedom.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-61574569729912391112016-10-17T01:22:18.880-07:002016-10-17T01:22:18.880-07:00Dave Giles, i have 9 independent variables and 1 d...Dave Giles, i have 9 independent variables and 1 depended variables, (total 10) i would it be possible to use ARDL Model?<br />amkhttp://www.blogger.com/profile/15867500960790926409noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-60544870865787543442016-10-15T10:00:40.022-07:002016-10-15T10:00:40.022-07:00I really had a hard time trying to understand char...I really had a hard time trying to understand characteristic roots until now. Your explanations were so much clearer. Thank you Prof., please keep this going! Hu Shanxiongnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-83200036990021310862016-10-12T04:23:12.969-07:002016-10-12T04:23:12.969-07:00I am sorry, but I also can't understand this e...I am sorry, but I also can't understand this example:<br />Pearson's product-moment correlation coefficient for sample data:<br />corr(x,y)= Σ[(Xi - X*)(Yi - Y*)] / {[Σ(Xi - X*)2][Σ(Yi - Y*)2]}1/2<br /><br />u*=(1+0)/2=0.5<br />v*=(1/sqrt(2)+1/sqrt(2))/2=1/sqrt(2)<br />w*=(0+1)/2=0.5<br />Then corr(u,v)=[(1-0.5)*{1/sqrt(2)-1/sqrt(2)}+(0-0.5)*{1/sqrt(2)-1/sqrt(2)}]/<br />/ [{(1-0.5)^2+(0-0.5)^2}*{{1/sqrt(2)-1/sqrt(2)}^2+{1/sqrt(2)-1/sqrt(2)}^2}]^(1/2) = 0<br />corr(v, w)=0<br />corr(u,w)=[(1-0.5)*(0-0.5)+(0-0.5)*(1-0.5)] /<br />/ [{(1-0.5)^2+(0-0.5)^2}*{(0-0.5)^2+(1-0.5)^2}]^(1/2)=<br />[-0.25-0.25]/[0.5*0.5]^(1/2)=-1<br />This result seems obvious to me. So, there is no correlation between u and v and between v and w, but corr(u,w)=-1.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-56252620243174699142016-10-11T16:54:54.668-07:002016-10-11T16:54:54.668-07:00Unless I am confused, I think that the K matrix de...Unless I am confused, I think that the K matrix defined after equation (2), like this:<br /> kij = 1 (if i = j, or j = i + 2); = -2 (if j = i + 1); = 0 (otherwise).<br /><br />should rather be so:<br /><br /> kij = -1 (if i = j); = 1 (if j = i + 1); = 0 (otherwise).<br /><br /> this is to get a vector containing: [y2-y0; y3-y1; ... ; yT-yT-2] and then get the normo of that with the (Kτ)' (Kτ)<br /> please do correct me if I am wrong.Unknownhttp://www.blogger.com/profile/00165740005837120968noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-52687413432022976462016-10-06T06:15:33.591-07:002016-10-06T06:15:33.591-07:00Just among the I(1) variables.Just among the I(1) variables.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-4755010303902384412016-10-02T20:43:28.497-07:002016-10-02T20:43:28.497-07:00We need more articles Dave, one a week???We need more articles Dave, one a week???not trampishttp://nottrampis.blogspot.com.au/noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-86573438085006182252016-09-30T09:28:13.156-07:002016-09-30T09:28:13.156-07:00You don't need normality.You don't need normality.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-35614482999753870082016-09-30T08:29:53.380-07:002016-09-30T08:29:53.380-07:00Good day sir, I run an ecm regression but am face ...Good day sir, I run an ecm regression but am face with problem of non normality of error term. How can I resolve the problem.Unknownhttp://www.blogger.com/profile/06925807378996458471noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-35724157690563528872016-09-29T10:17:29.823-07:002016-09-29T10:17:29.823-07:00Dear Sir,
I am working with 5 variables where only...Dear Sir,<br />I am working with 5 variables where only two of the independent variables are I(0) and all the rest are I(1). Can i run a VECM here(assuming cointegration exists)? For VECM to be operational, variables must be cointegrated so wheteher i should search for cointegration among all[I(0) and I(1)] the variables or only among the I(1) variables. Thank You in advance for your help. <br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-70531764836235501692016-09-28T16:31:22.420-07:002016-09-28T16:31:22.420-07:00Thanks!Thanks!Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-49439672168834370862016-09-28T14:11:43.568-07:002016-09-28T14:11:43.568-07:00Given your interest in breakpoints and Bayesian me...Given your interest in breakpoints and Bayesian methods, this article should be of interest:<br /><br />http://www.sciencedirect.com/science/article/pii/S0167947315002935Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-67757281910261883482016-09-28T11:28:59.834-07:002016-09-28T11:28:59.834-07:00It's simply that there are 2 conventions. Then...It's simply that there are 2 conventions. Then you either look at the roots, or the inverse roots. In the end it amounts to the same thing.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-3537983258168956802016-09-28T07:45:35.016-07:002016-09-28T07:45:35.016-07:00Dear Professor, thanks a lot for sharing this info...Dear Professor, thanks a lot for sharing this information. I have a doubt in the second characteristic equation (zp - γ1 zp-1 - γ2 zp-2 - ...... - γp = 0) of the AR model. The signs of the terms in the equation are different from what I observed in other texts (zp + γ1 zp-1 + γ2 zp-2 + ...... + γp = 0). Of course, the change in signs have significant impacts on assessing the behaviour of the process. Could you please share your comments on this?Ramesh Perumalhttp://nthu.academia.edu/RameshPerumalnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-25424577398568270712016-09-27T12:12:42.667-07:002016-09-27T12:12:42.667-07:00Brian - thanks very much. Fixed. DGBrian - thanks very much. Fixed. DGDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-43690413250702095852016-09-27T06:05:46.858-07:002016-09-27T06:05:46.858-07:00The post applies to both situations.The post applies to both situations.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.com