observations. This rate is held constant in Poisson, but varies in
Negative Binomial.
Can we interpret the value of lambda literally, i.e. lambda = 1 --> one
event per time period, lambda = 10 --> ten observations per time period,
etc? I guess we need to understand this to select an appropriate range
for lambda..
Thanks,
Stan
****************************
Stanislav Markus
Ph.D. Candidate
Harvard University
Department of Government
e: smarkus(a)fas.harvard.edu
t: 617.513.5407
-----Original Message-----
From: gov2001-l-admin(a)fas.harvard.edu
[mailto:gov2001-l-admin@fas.harvard.edu] On Behalf Of Kosuke Imai
Sent: Wednesday, February 26, 2003 11:27 AM
To: Olivia Lau
Cc: gov2001-l(a)fas.harvard.edu
Subject: Re: [gov2001-l] Problem 4
You are right that gamma is a continuous distribution. So, you have to
discretize it by creating a sequence of values for an appropriate range
of
lambda. Then, you can create a matrix (in R) where its i,j element
represents the joint density evaluated at the ith value of Y and the jth
value of lambda. Now, if you sum over the column (or "scoop" to use
Gary's
word), you will obtain the marginal density of Y. Is that clear now?
Kosuke
On Wed, 26 Feb 2003, Olivia Lau wrote:
> I think that the confusion is this: If lambda is distributed
> according to the gamma (continuous) distribution, how can we
> "construct a matrix"? Do you really want us to construct an
> actual matrix (in R), for which (I think) we would need to
> evaluate the poisson distribution at specific values of lambda,
> or do you just want us to do it theoretically and integrate to
> get the marginal density over some range of y?
>
> Am I still lost?
>
> Thanks, Olivia.
>
> ----- Original Message -----
> From: "Kosuke Imai" <kimai(a)fas.harvard.edu>
> To: <gov2001-l(a)fas.harvard.edu>
> Sent: Wednesday, February 26, 2003 9:57 AM
> Subject: [gov2001-l] Problem 4
>
>
> > A number of people have asked questions about problem 4 last
> night. Here
> > is a basic idea. You want to evaluate the "joint" density with
> different
> > values of y and lambda where y is poisson and lambda is gamma.
> This is the
> > joint density function, so you are not drawing random
> variables from any
> > distribution. you are simply evaluating the height of the
> density. To do
> > this, you need to create a matrix where its row and column
> represents y
> > and labmda, respectively. Then, to integrate out lambda, you
> want to sum
> > over the column. After normalizing it, this gives you the
> marginal
> > density of negative binomial.
> >
> > Kosuke
> >
> > _______________________________________________
> > gov2001-l mailing list
> > gov2001-l(a)fas.harvard.edu
> > http://www.fas.harvard.edu/mailman/listinfo/gov2001-l
> >
>
> _______________________________________________
> gov2001-l mailing list
> gov2001-l(a)fas.harvard.edu
> http://www.fas.harvard.edu/mailman/listinfo/gov2001-l
>
_______________________________________________
gov2001-l mailing list
gov2001-l(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov2001-l
Hello,
I have a question about implementing regressions in R with panel corrected
standard errors with a correction for autocorrelation between the error
terms. The reason I am reaching out to you is because I came across a
resource online dated from a while back
<https://lists.gking.harvard.edu/pipermail/gov2001/2008-April/001938.html>
when you indicated "We are still working on a function that does PCSE AND
AR(1) correction AND works for unbalanced panels (right now we only have
the first two)."
My question is, what progress has been made on such a program/package in R?
My goal is to adapt an analysis I performed in Stata that took the form of
"xtpcse [depvar] [indvars], corr(ar1)". When I used the pcse package in R,
I received an error message stating that "Length of groupN and groupT must
equal nrows of using data", which I believe indicates that my panels were
unbalanced.
Sorry for contacting you out of the blue, but I thought maybe you knew of a
quick solution given your post in the hyperlink above.
Thanks.
Ben
Dear Users,
I'm using the Zelig package in R to predict the HIV prevalence. In others words, I want to estimate the model that regresses the HIV status (HIV+ or HIV-) on the sex (male, female).
I have a problem of getting the residuals and the predicted values of the models with and without covariates respectively :
resu1 <- zelig(HIV~1,model = "logit.survey",weights = ~Weight,data=PREVATOUT)
resu2 <- zelig(HIV~Sex,model = "logit.survey",weights = ~Weight,data=PREVATOUT)
where HIV is a dichotomous indicator (HIV=0 or HIV=1), Weight is the survey sampling weight, Sex (Sex=1 for men and Sex=2 for female) and PREVATOUT is my dataset containing other covariates. How can I get the residuals and the predicted values of the models?
Any idea?
Thanks.
***************************************
Emmanuel BARANKANIRA
Phone : (+33)623575806
E-mail : baremma2002(a)yahoo.fr
France
***************************************
Hi everyone,
Someone left their black umbrella in section tonight. If that was your
umbrella -- let me know -- I have it.
Best,
Molly
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
https://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hello Gov 2001 Alumni!
I hope everyone has had a relaxing summer and is enjoying what’s left of it!
I am the new graduate student coordinator for the Applied Statistics
Workshop (Gov 3009) at IQSS and would like to invite all of you to attend
the workshop. The workshop features a multidisciplinary forum for presenting
research with statistical innovations and applications. Starting with
Wednesday, Sept. 7, we will meet every Wednesday from 12-1:30 pm in
CGIS-Knafel 354 (1737 Cambridge Street). As always, lunch will be provided.
Please note that you don’t have to formally enroll in the workshop to
attend. The schedule for the semester will be available shortly at
http://www.iq.harvard.edu/events/node/1208. Furthermore, if you would like
your name to be added to the mailing list, please let me know.
My apologies to those who have received this announcement through the gov
grads listserv also.
Thank you,
Konstantin
--
Konstantin Kashin
Ph.D. Student in Government
Harvard University
Mobile: 978-844-0538
E-mail: kkashin(a)fas.harvard.edu
Site: http://people.fas.harvard.edu/~kkashin/
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
https://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hello to anyone who is still reading this,
I am working with elections data in which I have the votes a candidate
received but not if they won or not. I want to add a couple columns
to my dataset just noting if the candidate won and by how much. I feel
like this should be easy. And yet…
Here is what I've done
length(unique(data$constituency))
mat<-as.vector(unique(data$constituency))
for(i in 1:229){
sub<-data[data$constituency==mat[i],]
ordered<-sub[order(sub[,"vote"], decreasing=TRUE),]
win<-ordered[1,]
second<-ordered[2,]
totalvotes<-sum(sub[,"vote"])
winvote<-win[,"vote"]
secvote<-second[,"vote"]
margin<-winvote-secvote
pcntmarg<-as.numeric(margin/totalvotes)
data[data[,"name"]==win[,"name"],]$marginwin[]<-pcntmarg
data[data[,"name"]==win[,"name"],]$elected[]<-1
}
Resulting in these errors:
Error in `$<-.data.frame`(`*tmp*`, "marginwin", value = numeric(0)) :
replacement has 0 rows, data has 1
Any words of wisdom?
Thanks from Ghana!!
Julie
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi Gang, here's an interesting project for anyone interested in health
policy. Please contact Lory directly if you're interested.
Gary
--
*Gary King* - Albert J. Weatherhead III University Professor - Director,
IQSS - Harvard University
GKing.Harvard.edu <http://gking.harvard.edu/> - King(a)Harvard.edu -
@kinggary<http://twitter.com/kinggary>- 617-500-7570 - Asst 495-9271 -
Fax 812-8581
---------- Forwarded message ----------
From: Lory Wallach <lwallach(a)firstopinion.com>
Date: Fri, May 20, 2011 at 12:27 PM
Subject: Re: First opinion-- consulting assistance.
To: Gary King <king(a)harvard.edu>
Hi Gary
The First Opinion is a medical knowledge-based system designed to give
medical advice to the general public and decision support to clinicians. The
System is web-based and can be accessed through any internet browser, smart
phone or dumb phone through interactive voice response and automatic
speech recognition
technologies in most languages. Using an Internet based authoring
approach; the First Opinion system encodes a highly useful core of expert
and general practitioner diagnostic and treatment knowledge into a computer
system that can be accessed by clinically trained individuals as well as by
any individual consumer. The System has two different front ends that
support both of these audiences. The System has the ability to consider
every disease of mankind (24,000) for anyone seeking consultation
For patients-- this provides free and easy access to diagnosis-- and answers
the questions-- what is wrong with me... what is my next best step for care,
was the diagnosis made by my provider correct.
For providers-- it acts as a clinical decision support system especially for
those that do not have access to that through an EMR.
We are seeking consulting help to size the global market. We know it is
huge-- and we want to put some measures around that. This could include what
is the gap in the global capacity to make a medical diagnosis, what is the
demand for receiving a correct and timely diagnosis, what is the cost of
incorrect or delayed diagnosis, what is the potential reduction in pain and
suffering, what is the impact on freeing up capacity so that health care
resources can be used more effectively?
Thank you for sharing this request for consultative services with your
capable students and colleagues.
Best,
Lory Wallach
858-733-2145
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi Class,
Sorry for the harassment far after the end of class. I'm using survey data with
weights and, if I use matching and readjust my quantity of interest to the
average treatment effect on the treated, should I still be using survey
weights? My inclination is that the answer is no, especially since I might be
using the weights from matching in the regression.
Any help is much appreciated!
Best,
Leslie
p.s. if any of you have ever used "svydesign" in R, I'm trying to figure out
what exactly i'm supposed to put for "id"...
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
--
gov2001 mailing list served by Harvard-MIT Data Center
List Address: gov2001(a)lists.gking.harvard.edu
Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=gov2001
Gov 2001,
I've been asked by the Extension School to remind all extension school
students that evaluations are due on Monday. Please fill them out they are
extremely important to us!
(I am pretty sure everyone else has an extra week, but be sure to do them!)
Brandon
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
IQSS is sponsoring what looks like will be a great conference on political
economy and the law. in case you're interested, pls see the attached.
Gary
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l