[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: test
At 03:28 PM 5/31/2004 +0200, email@example.com wrote:
According to the reference manual, the test was suggested by Pregibon
(1979) based on an idea of Tukey (1949). It also says that Pregibon (1980)
suggested a slightly different method that has come to be known as
"Pregibon's goodness-of-link test" - but Stata uses the older version
because it is universally applicable. So, I'm not sure if there is a name
for the test, but there is a source for it.
what is the real name of the test called "linktest" in Stata?Is there an
Collin reports 2 sets of numbers that are both called the condition
index. I'm not sure if they have separate names or not. The difference is
in the estimation, i.e. with index 1 variables are centered about their
means first and in index2 they are not. My intuition leads me to favor
approach 1 but it turns out there is a lot of controversy on this. Belsley
Kuh and Welsch discuss this in their 1980 book on regression diagnostics.
And the real name of the Condition Index1 (collin)???
Many many thanks...
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
* For searches and help try:
- st: test
- From: firstname.lastname@example.org