Let
M = sample size
pi = fraction of hits
N = Binomial sample size
The last two are just parameters for the model: Binomial(pi,
N).
For i in 1:M, take a rbinom(i, pi, N) and calculate the sample
mean and store it. Then plot 1:M against your vector of sample
means. Do this for each of the specified cases, and remember to
constrain your axes to the same dimensions in both directions.
Include a line representing the converged distribution, and
another set of lines representing the 95% CI (CI == CLT CI, not
quantile()).
----- Original Message -----
From: "Jens Hainmueller" <jens_hainmueller(a)ksg05.harvard.edu>
To: <gov2001-l(a)lists.fas.harvard.edu>
Sent: Saturday, October 16, 2004 11:48 AM
Subject: AW: [gov2001-l] plot sample means against sample size?
what is meant by question 3 "Plot sample means
against sample
size?" If we
take many iid draws from the binominal holding the sample size
N constant at
either (10,40, or 100) as the queston seems to ask for, there
will be no
variation in N for each experiment. So why plot the simulated
means against
N if N doesn't vary?
any help is greatly appreciated. thanks!
best,
jens
-----Ursprungliche Nachricht-----
Von: gov2001-l-bounces(a)lists.fas.harvard.edu
[mailto:gov2001-l-bounces@lists.fas.harvard.edu]
Gesendet: Saturday, October 16, 2004 11:34 AM
An: gov2001-l(a)lists.fas.harvard.edu
Betreff: [gov2001-l] Inverse Chi distribution
Hi all,
I've been looking at problem one of this week's homework,
and I don't
understand what the Inverse Chi squared distribution is. I
understand the
traditional chi squared distribution f(v) (Rice pg 178), but
that uses
degrees of freedom rather than sigma squared. Could someone
explain the
inverse chi squared or suggest a reference for where to look.
I've tried
both King and Rice and haven't found anything. I've searched
online and
found some references to inverse chi squared distributions,
but they seem
to parameterize it differently than is suggested in the
homework.
Thanks,
Ellie
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l