Sorry folks, but I need to add two errata and one further
consideration to my earlier post, if only to help out other
Statalisters who may wish to contribute:
(E1) Either I have collected some data for this or I have not: the
answer, as you all probably guessed, is that I have;
(E2) The last refererence should, in fact, read: Stimson JA (1994)
"Regression in Space and Time: A Statistical Essay", American Journal
of Political Science 29 (4): 914-47, and _not_ 1971 as I printed!
(C1) I have perhaps have not thought as hard about the use of lagged
variables as I should have done for these (at the moment, theoretical)
models. I should give here a very brief explanation as to what these
WFERs actually are. Essentially, these are based on international
chess ratings devised by the Hungarian-American physics lecturer (and
chess master), Professor Arpad Elo (1903-92). Incredibly, some bloke
with bags of SQL skills (and lots of time), has tweaked Elo's original
formula and calculated ratings for every single national football team
in the world from their very first game onwards (for more, see:
http://www.eloratings.net/world.html). Given the way it is calculated,
a team's 'past form' is already built into their current WFER, so I am
unsure as to whether lagged variables would be useful for my models
here. I guess this is another question for the list. If I were an
economist, I would be in three minds about it!
Ta very much, once again.
--
Clive Nicholas
[Please DO NOT mail me personally here, but at
<[email protected]>. Thanks!]
"Courage is going from failure to failure without losing enthusiasm."
-- Winston Churchill
*
* 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/
