Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Scott Merryman <scott.merryman@gmail.com> |
| To | statalist@hsphsun2.harvard.edu |
| Subject | Re: st: white cross-section standard errors |
| Date | Mon, 7 Nov 2011 04:15:46 -0600 |
A Google search of "white diagonal standard errors" brought up this page: http://www.econ.canterbury.ac.nz/personal_pages/bob_reed/Courses/ECON_324/Week7/Assignment_Week7.pdf " Another difference is that Stata does not have as many options for specifying the structure of the error variance-covariance matrix as Eview does. When estimating OLS, there are only three options: (i) conventional OLS standard errors, (ii) “robust” standard errors -- which is called “White diagonal” standard errors in Eviews; and (iii) “Clustered robust” standard errors -- which is very close to what EViews calls either “White period” or “White cross-section” standard errors, depending on whether the “cluster” variable is the cross-sectional (e.g. firm) or time series (e.g. year) variable" Scott On Mon, Nov 7, 2011 at 1:20 AM, Mohammad Reza Farzanegan <Mohammad_Reza.Farzanegan@mailbox.tu-dresden.de> wrote: > Thank you for your suggestions. It is however still not clear to me which > option(s) in a regression exactly lead to for example to "white cross > section" or "white period", "white diagonal" standard errors in STATA. > > >> Mohammad Reza Farzanegan <Mohammad_Reza.Farzanegan@mailbox.tu-dresden.de>: >> There are many varieties of robust standard errors in Stata; see the >> manual entry on -regress- for a few or see e.g. >> help xtpcse >> help xtgee##correlation >> ssc install ivreg2 >> help ivreg2 >> >> On Sun, Nov 6, 2011 at 7:54 AM, Mohammad Reza Farzanegan >> <Mohammad_Reza.Farzanegan@mailbox.tu-dresden.de> wrote: >>> >>> Dear statalist participants, >>> >>> As you may know, EViews has 7 options to calculate the "robust" standard >>> error, while as far as I know, in STATA one can only use the "robust" >>> option. >>> >>> Now my question: Do you know a do-file in STATA to calculate white >>> cross-section standard errors ( or the other options such as white >>> period, >>> white diagonal and so on)? >> * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/