tag:blogger.com,1999:blog-2198942534740642384.post6931363994570078464..comments2023-10-24T03:16:41.009-07:00Comments on Econometrics Beat: Dave Giles' Blog: Allocation Models With Bounded Dependent VariablesDave Gileshttp://www.blogger.com/profile/05389606956062019445noreply@blogger.comBlogger10125tag:blogger.com,1999:blog-2198942534740642384.post-85657664795413113142013-09-14T11:03:23.481-07:002013-09-14T11:03:23.481-07:00Thank you very much! How does the model change whe...Thank you very much! How does the model change when we are in a panel data structure, i.e. for the same year and the same country the sum of the independent variables across the equations is 1?<br />Do I have to perform an OLS estimation or something else?Pieronoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-4451205256062885322013-07-25T21:14:51.651-07:002013-07-25T21:14:51.651-07:00There is an expository paper discussing the topic ...There is an expository paper discussing the topic for panels. We use the approach fairly often for time series analysis of allocation data. "Cake Slicing and Revealed Government Preference" Bell Journal of Economics, 1982<br />Aaronhttps://www.blogger.com/profile/15160075483092543629noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-63636538045101271252013-07-16T13:46:13.091-07:002013-07-16T13:46:13.091-07:00I see..ok, thanks, I'll give it a shot!I see..ok, thanks, I'll give it a shot! Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-34548554748256604152013-07-16T07:47:34.166-07:002013-07-16T07:47:34.166-07:00Hi - no idea, as I'm not a STATA user. It'...Hi - no idea, as I'm not a STATA user. It's not in EViews either, but it's trivial to code up the log-likelihood function with 3 lines of code, as in my EViews workfile:<br /><br />@logl ll1<br />mu1=exp(alpha(1)+beta(1)*x )/ (1 + exp(alpha(1)+beta(1)*x))<br />ll1=log(@gamma(phi(1)))-log(@gamma(phi(1)*mu1))-log(@gamma(phi(1)*(1-mu1)))+(phi(1)*mu1-1)*log(y1)+(phi(1)*(1-mu1)-1)*log(1-y1)<br /><br /><br />and I know you can do the same sort of thing easily in STATA.<br /><br />DGDave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-13024196021557093882013-07-16T06:07:07.986-07:002013-07-16T06:07:07.986-07:00Do you know if there is a code for Beta regression...Do you know if there is a code for Beta regression for STATA? I'm using the Papke and Woolridge GLM QMLE method with a logit link function, but would like to also try the Beta regression for proportional data. Particularly since I believe the Beta regression would be more applicable to smaller samples, right? Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-18110348681064153762013-07-08T08:52:09.267-07:002013-07-08T08:52:09.267-07:00One concern would be lack of independence. Indepen...One concern would be lack of independence. Independence is assumed in the construction of the log-likelihood function used here. However, if any dependencies were incorporated properly into the specification of the likelihood function, then no problem. Dave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-62567407498344421602013-07-08T06:10:53.160-07:002013-07-08T06:10:53.160-07:00Any advice if the ratio's are a time series? ...Any advice if the ratio's are a time series? What are your thoughts on including lags of each of the components as regressors with various these link functions?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-49897821289432550592013-07-06T15:17:06.992-07:002013-07-06T15:17:06.992-07:00No worries, thanks for the quick fix!No worries, thanks for the quick fix!Achim Zeileishttp://eeecon.uibk.ac.at/~zeileis/noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-20327051030235363622013-07-06T12:27:41.138-07:002013-07-06T12:27:41.138-07:00Achim - my apologies!! Fixed now.
DGAchim - my apologies!! Fixed now.<br />DGDave Gileshttps://www.blogger.com/profile/05389606956062019445noreply@blogger.comtag:blogger.com,1999:blog-2198942534740642384.post-70691923391307495962013-07-06T12:14:29.318-07:002013-07-06T12:14:29.318-07:00Thanks for the nice post and the publicity for our...Thanks for the nice post and the publicity for our R package. Note, however, that the JSS manuscript was co-authored by Francisco and myself, not Silvia.Achim Zeileishttp://eeecon.uibk.ac.at/~zeileis/noreply@blogger.com