Hi Everyone,
I've also decided to extend the the due date for problem set 7. It is
now due at 4:00 pm on Tue. Nov. 30. Problem set 7 should be handed in
either in section, or in Ryan's CBRSS Room 31 mailbox no later than
4pm, Tuesday, 30 November.
Have a good Thanksgiving.
Best,
Kevin
------------------------------------------------------
Kevin Quinn
Assistant Professor
Department of Government and
Center for Basic Research in the Social Sciences
34 Kirkland Street
Harvard University
Cambridge, MA 02138
Hi Everyone,
I've decided to move the due date for the final paper. The paper is
now due at 5:00 pm on January 3. The papers should be submitted
electronically to Alison, Ryan, and me.
I've revised the syllabus to reflect this change and have put the new
version up on the course website.
Best,
Kevin
------------------------------------------------------
Kevin Quinn
Assistant Professor
Department of Government and
Center for Basic Research in the Social Sciences
34 Kirkland Street
Harvard University
Cambridge, MA 02138
Does anyone know offhand if the sci. ctr. lab is open on Thanksgiving? I don't
seem to be able to find their hours and the help desk is closed right now....
Thanks!
Marie
Can anyone tell me why we need to regress prestige on income alone in
this problem? They tell us to do it, and then don't mention it for the
rest of the problem.
andy
Hi Everyone,
Revised lecture notes are now up on the course website.
Best,
Kevin
------------------------------------------------------
Kevin Quinn
Assistant Professor
Department of Government and
Center for Basic Research in the Social Sciences
34 Kirkland Street
Harvard University
Cambridge, MA 02138
Greetings,
I am having trouble getting the dimensions of the x and y vector to
conform. I think my problem is that I am not getting the barro data clean
of all the na's. However, I have done the process three times, and I
can't seem to get it right. any suggestions?
Regards,
Sheldon
Hi Everyone,
I put some Monte Carlo examples coded in R up on the course website
under Handouts > R_Examples. The files are:
betaCI.R
betaSampDist.R
FSampDist.R
IncVarBias.R
tSampDist.R
I don't think I'll have time to go through these tomorrow in lecture,
but hopefully things are commented well enough that you can look at
the code, see what is happening, and then source() the file in to R to
run it and observe what actually does happen.
Monte Carlo examples can help you build some intuition about what is
going on with frequentist inference since with a computer we really
can take repeated samples from a population and look at the sampling
distribution of our estimators and test statistics. If you have
questions about what is going on with any of these programs feel free
to send the questions to the course listserv.
Best,
Kevin
------------------------------------------------------
Kevin Quinn
Assistant Professor
Department of Government and
Center for Basic Research in the Social Sciences
34 Kirkland Street
Harvard University
Cambridge, MA 02138
For those who have done Problem 4 - can you help explain what the
question is asking for ?
Specifically, I don't really understand what part (a) is asking for -
isn't the partial correlation between education & income exactly what we
calculated in the previous exercise - where education = Y and income =
X1, controlling for X2 and X3, proportion under 18 and proportion urban?
And, as to Problem 3(a), I'm not sure whether my slope coefficient is
the same as Problem 2.
In problem 2, I got a slope for income = 0.0806.
In problem 3, I got a slope = 8.07e^(-2).
Are these the same? I don't know what e is (I remember vaguely, from
math camp, that it refers to really really small numbers, epsilon
balls??), but don't know what it translates to, here.
Thanks !
Brodi
Having some trouble with getting the abline. (sorry-it's been a very long day
:) ) I get a lattice plot with the following:
library(car)
data(Anscombe)
attach(Anscombe)
anscombe.mod1<-lm(education~young+urban)
anscombe.mod2<-lm(income~young+urban)
res1<-anscombe.mod1$residuals
res2<-anscombe.mod2$residuals
reg3<-lm(res1~res2, data=Anscombe)
plot(Anscombe)
abline(reg3)
The abline is shown in the lower right box where there are no data points...
Pointers???
Many, many thanks!
Marie