tag:blogger.com,1999:blog-2198942534740642384.comments2017-03-28T03:45:32.915-07:00Econometrics Beat: Dave Giles' BlogDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger3631125tag:blogger.com,1999:blog-2198942534740642384.post-18976124511494437102017-03-25T09:59:26.894-07:002017-03-25T09:59:26.894-07:00That's correct.That's correct.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-46095505610369947942017-03-25T09:46:45.432-07:002017-03-25T09:46:45.432-07:00Dear prof. Giles,
Thank you for all your valuable...Dear prof. Giles,<br /><br />Thank you for all your valuable contributions. This blog has teached me a lot.<br /><br />Still, I have one question regarding TY causality. Does it refer to short run or long run causality?<br /><br />My best guess is short run, the same as "regular" granger causality. As opposed to the causality through the error correction term (in VECM) that may be considered as long run causality.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-90722264672027167502017-03-25T08:59:10.930-07:002017-03-25T08:59:10.930-07:00Professor, check out this funny website that gener...Professor, check out this funny website that generates seriously meaningless paragraphs related to philosophy:<br />http://www.elsewhere.org/journal/pomo/Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-16258569100700424222017-03-20T13:05:08.008-07:002017-03-20T13:05:08.008-07:00Thanks - I wasn't aware of this, and it looks ...Thanks - I wasn't aware of this, and it looks great!Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-19309369110155074652017-03-20T08:33:39.518-07:002017-03-20T08:33:39.518-07:00Dave,
You might be interested to know that there ...Dave,<br /><br />You might be interested to know that there is an undergrad version of this competition, now in its third year, at the University of Chicago: https://www.facebook.com/events/203669573447162/<br /><br />Sign-up closes April 1.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-61644167378575396082017-03-20T07:37:24.182-07:002017-03-20T07:37:24.182-07:00Any version of EViews will open them. You can also...Any version of EViews will open them. You can also view them with any text editor - e.g., notepad.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-78580543144381837792017-03-20T03:22:48.346-07:002017-03-20T03:22:48.346-07:00It is a very helpful book for practitioners as wel...It is a very helpful book for practitioners as well as learners. How can we access the .prg files that have been used in this book? It says it is in Eviews content folder. Is it available only for Reviews 9.5?Abhishek kumar rohithttp://www.blogger.com/profile/16426166455480279958noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-75811922153117371432017-03-19T14:41:10.852-07:002017-03-19T14:41:10.852-07:00No, not this time -but Monash was the winner a few...No, not this time -but Monash was the winner a few years back.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-18241973092297597582017-03-19T14:20:34.837-07:002017-03-19T14:20:34.837-07:00no uni from Australia!no uni from Australia!not trampishttp://nottrampis.blogspot.com.au/noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-82287969667905612742017-03-19T09:21:19.011-07:002017-03-19T09:21:19.011-07:00Thank U Prof.D Giles!I appreciate. Thank U Prof.D Giles!I appreciate. Kene Offornoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-12958487375930078842017-03-16T05:30:16.311-07:002017-03-16T05:30:16.311-07:00another question:
Suppose you estimate
y_i = b_...another question:<br /><br />Suppose you estimate <br /><br />y_i = b_0 + b_1x_1_i + b_2x_2_i (1)<br /><br />Regress<br /><br />e_i=z_0 + z_1_i (2)<br /><br />where e is the residual of a regression of y on a constant and x_2.<br /><br />Show |z_1| is smaller/equal |b_1| and<br />how to change (2) such that z_1=b_1Unknownhttp://www.blogger.com/profile/04874901116159051943noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-43033131952726544392017-03-16T03:31:36.188-07:002017-03-16T03:31:36.188-07:00Hello Professor Giles.
Thanks for your work: it is...Hello Professor Giles.<br />Thanks for your work: it is useful to people from all over the world!<br />This question is very interesting for me: it seems that the OLS approach for the estimation of the parameters of the long-run model is not the only one. For example, Pesaran Shin and Smith (1999 and 2001) tell about an "ARDL approach" to estimating the parameters of the long-run model. <br />Your opinion in this regard would be much appreciated.<br />Andreahttps://www.facebook.com/andrea.molentnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-60320490769071494142017-03-11T20:30:10.035-08:002017-03-11T20:30:10.035-08:00My pleasure - thanks for the helpful discussion.
D...My pleasure - thanks for the helpful discussion.<br />DGDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-64046697080893615782017-03-11T20:25:14.959-08:002017-03-11T20:25:14.959-08:00Yes, everything makes perfect sense now. When doin...Yes, everything makes perfect sense now. When doing these simulations, I did not fully appreciate the importance of having fixed regressors because I would usually select a large n. For example, I often set my n=500 (sample size) and N=5,000 (number of replications). Thus, in many cases, I would not reject normality of the sampling distribution; however, as you clearly point out (using the dynamic OLS model as an example), with smaller sample sizes, the distribution of OLS estimates are not normally distributed. <br /><br />When I set n=10 and N=5,000, I did one simulation where I drew x from a normal(0,1) each time I drew the error term, and I was able to reject normality of the 5,000 OLS coefficient estimates very easily; the p-values were extremely small using a test similar to the Jarque-Bera test. I then did another simulation where I drew x once and kept it fixed across all the replications; I could no longer reject normality. The p-value was close to 0.56.<br /><br />Thus, I can clearly see the important of keeping x as a fixed regressor when trying to do simulations where one is interested in computing empirical power, for example. Thanks again!Chrishttp://www.blogger.com/profile/00614012660427960869noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-11510193660724885012017-03-11T18:56:28.345-08:002017-03-11T18:56:28.345-08:00Chris - if the random regressors are uncorrelated ...Chris - if the random regressors are uncorrelated with the errors, then of course there is no bias. Does that make sense?<br /><br />DavidDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-54405717687612514832017-03-11T12:51:35.042-08:002017-03-11T12:51:35.042-08:00Thanks for the reply. I just want to clarify one a...Thanks for the reply. I just want to clarify one additional question I have: In econometrics, the assumption of non-stochastic regressors is often relaxed. Thus, the vector of regressors may be viewed as either all stochastic or a mixture of stochastic and non-stochastic regressors. I performed a simple simulation with the following data generating process:<br />y = \beta_0 + \beta_1 x + \epsilon,<br />where \beta_0 = 1, \beta_1 = 0.5, and x and \epsilon are normally distributed with mean 0 and variance 1. I draw a sample of 500 observation on x and \epsilon, compute y, and estimate the OLS regression coefficients. I then repeat this exercise 1,000 times and collect the results---there is no assumption of x being fixed in repeated samples. OLS, on average, still demonstrates that the parameter estimate for \beta_1 is unbiased. May I take this to assume that I have demonstrated conditional unbiasedness? Then one could argue that I could integrate over the conditional distribution to demonstrate unconditional bias. Does this line of thought hold? I want to think how this finding connects with the idea you have suggested above, which suggests that using stochastic regressors will allow me to see bias introduced into the OLS estimator. Thank you again for your insightful blog and comments. Chrishttp://www.blogger.com/profile/00614012660427960869noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-10045304830659200202017-03-11T09:05:08.972-08:002017-03-11T09:05:08.972-08:00Absolutely essential for this exercise. Unless you...Absolutely essential for this exercise. Unless you want to see the bias introduced into the OLS estimator when you have random regressors.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-72398311049617352712017-03-11T09:02:23.077-08:002017-03-11T09:02:23.077-08:00Dear Prof. Giles,
Thank you for these posts about...Dear Prof. Giles,<br /><br />Thank you for these posts about using Monte Carlo simulation to illustrate properties of the sampling distribution of OLS parameters. I have a question about the necessity of keeping the observed values of the regressors, x2 and x3, as non-random. How necessary is it to keep these regressors "fixed in repeated samples"?Chrishttp://www.blogger.com/profile/00614012660427960869noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-84396999997697767842017-03-05T10:07:56.418-08:002017-03-05T10:07:56.418-08:00Thanks - that helps. I don't think there is an...Thanks - that helps. I don't think there is any compelling reason to include a (deterministic) trend.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-44904172801613540022017-03-05T09:42:19.590-08:002017-03-05T09:42:19.590-08:00I was thinking in terms of time series with unit r...I was thinking in terms of time series with unit roots and/or varying degrees of seasonality. If a time trend was included, then the beta space that we plot the coefficients in from the rolling regression could more confidently be attributed to changes in the explanatory variable rather than the trend (delta y = alpha if delta e =0). But I could be over-thinking it. If the window is small enough, perhaps the trend will not be justified. I realize some researchers set their windows at intervals to let each iteration have equal exposure to seasonality. So the trend component is typically not justified in a rolling regression?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-80414162756114985092017-03-05T07:41:53.116-08:002017-03-05T07:41:53.116-08:00Exogenous.Exogenous.Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-69535003145491511152017-03-05T07:40:40.923-08:002017-03-05T07:40:40.923-08:00I don't see what the intuition is...... do you...I don't see what the intuition is...... do you want to elaborate?Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-12796170551457695862017-03-05T06:48:26.393-08:002017-03-05T06:48:26.393-08:00Just a quick question, is it conventional to inclu...Just a quick question, is it conventional to include a trend component when conducting a rolling regression? In eviews terms, including @trend among the right hand side variables. I did not see this in the rolling regression literature, but I think there is intuition for including. Please advise, thank you.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-91133632154634904542017-03-05T05:27:32.422-08:002017-03-05T05:27:32.422-08:00Dear prof. Giles,
I have one basic question regar...Dear prof. Giles,<br /><br />I have one basic question regarding dummies.<br />Say I'm estimating Gregory & Hansen cointegation between Y and X and the regime shift model supports cointegration. Hence, in order to obtain long run equilibrium relationship between Y and X (using DOLS in Eviews) I have to include two dummy variables: a shift dummy and a slope dummy. The question is how should I include these dummies, as endogenous or as exogenous variables?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-69206609283646779002017-03-02T07:48:13.338-08:002017-03-02T07:48:13.338-08:00Fathy - no, it will be different, but I don't ...Fathy - no, it will be different, but I don't have any references offhand.<br /><br />Dave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.com