Also, does the phrase "all quantities of
interest" include standard errors
or confidence
intervals? That is, when we simulate the expected
annual number of events
under ppid = 1 and
ppid = 0 while holding sumil at its mean, do we need
to report standard
errors or confidence
intervals for the expected values?
It's fine if you report SE for the first difference only.
Does anyone know whether we're supposed to use
the rpoiss and rnbinom
functions when
simulating the expected values for the poisson and
negative binomial
distributions?
No. no need for these.
Hth,
jens
From: gov2001-l-bounces at
lists.fas.harvard.edu
[mailto:gov2001-l-bounces at
lists.fas.harvard.edu] On Behalf Of Sean Li
Sent: Sunday, April 20, 2008 4:12 PM
To: gov2001-l at
lists.fas.harvard.edu
Subject: Re: [gov2001-l] Question 2, Part C2
Correction: rpois, not rpoiss.
On Sun, Apr 20, 2008 at 4:10 PM, Sean Li <seanli at fas.harvard.edu> wrote:
Thanks!
Sean