For those of you who haven't used R before, I uploaded a brief
introduction to R which is written by Dan Ho for another class. It's under
``Handout'' on the course website.
A reminder that today I will hold an extra office hour right after the
lecture in my office at CBRSS. Tomorrow, I will have a regular office hour
from 4 to 6pm.
Kosuke
Here's our abstract. Any comments are greatly appreciated.
Best,
Dan and Colin
Crowded Cities, Crowded Streets: How Population Density Affects Local
Participation
Dan Hopkins and Colin Moore
Drawing on a survey of 29,000 Americans from communities throughout the United
States, this paper expands on J. Eric Oliver's (2000) argument that living in a
larger city decreases the probability that one will engage in a variety of
forms of local political participation. By shifting the causal emphasis from a
city¡¯s size to its population density, we illustrate that the effects of an
individual¡¯s environment depend crucially on the kind of participation in
question. As population density increases, an individual is 2.7% less likely
to attend a public meeting, but 2.0% more likely to participate in a
demonstration, boycott, or protest.
I will have a regular office hour today from 4 to 6pm.
Kosuke
---------- Forwarded message ----------
Date: Tue, 29 Apr 2003 11:58:30 -0400 (EDT)
From: Yevgeniy Kirpichevsky <kirpich(a)fas.harvard.edu>
To: Kosuke Imai <kimai(a)fas.harvard.edu>
Subject: office hrs
Kosuke,
will you be in the office today @4?
-y and p
When you run a linear regression that includes interaction terms, how do
you interpret the standard errors?
For instance, we have two dummy variables, one that indicates the presence
of a popular initiative on a ballot and another that indicates whether a
presidential race was held that year. So we get estimated coefficients
and standard errors for:
initiative
presrace
initiative*presrace
When initiative and presrace both = 1 (both present on the ballot), the
total predicted effect of the presence of an initiative equals the
coefficient on initiative plus the coefficient on initiative*presrace.
What is the standard error for this predicted effect? Do we add together
the standard errors for initiative + initiative*presrace, or do something
else?
Thanks!
Anna
Hi,
We are attempting to run a regression in "first differences" where the
functional form is the same as a normal regression, but all the variables
are "differenced" or have a delta in front of them.
Does this simply involve taking each variable in our dataset and doing:
DELTA Y = Y_t - Y_{t-1} ?
We have tried this and we are getting very different results from the article
we are trying to replicate. Before concluding that they did something wrong,
we wanted to make sure we are doing everything correctly. Is there some
additional trick to running a regression with first differences?
Thanks,
Phillip
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893 DORM
(617)851-8220 CELL
lipscy(a)fas.harvard.edu
http://www.people.fas.harvard.edu/~lipscy/
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
How can you extract the standard errors for the parameter estimates from a
fitted model (like model$coef for coefficients? There must be an easy way to
do this that I am overlooking.
Thanks,
JOhn.
Dear all,
Does anyone know of a paper that deals with a multiple equation
model with two different distributions? I think I have a case
where I have stochastically dependent Ys (probability of crisis,
length of crisis). The problem is that the first Y (pr crisis)
is logisitic and the second is weibull.
Any suggestions?
Thanks,
Olivia
I think I have an intuitive sense of the answer to this question, but it's come
up in one of the datasets we are trying to replicate so just thought I would
canvass others' opinions.
Suppose a variable such as "divided government" is coded -1 for unified, 0 for
partial, and 1 for divided. After running a regression, the variable is not
significant. Is it legitimate then to reassign values such as 0 for unified, 1
for partial, and 255 for divided to produce significant results?
My intuition tells me that this would be legitimate only if there is some prior
theoretical reason to believe that divided government has 255 times the effect
of partial. Is that right?
Thanks,
Phillip.
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893 DORM
(617)851-8220 CELL
lipscy(a)fas.harvard.edu
http://www.people.fas.harvard.edu/~lipscy/
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
Has anyone had success using the Windows version of amelia? I have been
unable to input my dataset; the program either fails to recognize the file
or crashes altogether. I have tried numerous formats, including Ascii,
txt, excel, comma delimited, and Stata without any luck.
Any suggestions would be greatly appreciated.
David
Hi everyone,
The two important things to be noted about the term paper.
1. Deadline: 4pm, May 14.
We decided to extend the due date to the end of the reading period, which
is May 14. There will be a box at the CBRSS first floor reception desk.
So, you (or your co-author) should turn in one hard copy of the paper in
the box. Note that we rarely give incompletes, so be sure to finish it by
the deadline.
2. Paper format:
The paper should follow *all* of the guidelines specified in
http://gking.harvard.edu/papers/papers.html Read this document carefully
as you write up your paper. We take this point seriously as we grade your
papers.
Good luck,
Kosuke