Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: {RE: [problem with Newey-West standard errors]


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: {RE: [problem with Newey-West standard errors]
Date   Sat, 11 Mar 2006 16:33:09 -0000

Please use an informative title for your posting. 

The report is in effect "it doesn't work" which 
is usually impossible to comment on, but more 
can be said in this case. 

I note in passing that you indicate, presumably 
schematically, that you typed 

newey indepvar depvar, lag(4) force 

However, the dependent variable would be named
first. 

On the key question, you can't have it both 
ways. It's as simple as that. -newey- won't 
function without a prior -tsset- and you can't 
-tsset- your data in terms of just a time 
variable, as it will detect repeated values, unless
you happen to have panel data for which at most
one member is recorded for each time point. 

If your question makes theoretical sense, 
and on that I can't comment, then the only 
way to do this with Stata would appear
to be by writing your own program.

Nick 
n.j.cox@durham.ac.uk 

Christine Dwane
 
> I need to have newey west standard errors, not in a panel 
> data context (i.e.
> setting the time and panel variables
> with "sort" and "tsset" commands), but with a pooled ols estimator.
> 
> So I try to do this by reading in the panel data set without 
> setting the time
> variable using tsset. I tried to use the command "newey 
> indepvar depvar, lag(4)
> force" but it won't work and I was just wondering if there is 
> another way to get
> the newey s.e. with this,
 

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2020 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index