How would we do first differences for part b?
So we have two scenarios, and the probabilities of 1, 2, 3 would change
from scenario 1 to scenario 2.
How do we capture these differences in a ternary diagram?
So would we have a matrix of three columns, first column being the
difference of probability of 1 for scenario 1 and 2, second column
being the difference of probability of 2 for scenario 1 and 2, etc.?
I guess my question is what exactly do we do for part b?
Thanks,
Chester
I tried to use the function "mvrnorm", and R returns that it could not find
that function. Any idea what I should do to get it?
yongwook
-----------------------------
Yongwook Ryu
PhD Student
Department of Government
Harvard University
Tel:617-493-3397
Email: yryu(a)fas.harvard.edu
-----------------------------
Dear All,
Quick question: in the data we are working with, people were asked how many
public meetings they had attended in the last year. The range of answers is
from 0 to 60, but the distribution across those values is spiked at the even
numbers (i.e. 612 people say 10 meetings and 634 say 12 meetings, while only 8
say 11 meetings, and 8 say 13 meetings). We were hoping to use the generalized
event count model, but were curious if that was a good choice given the data
and what seems to be its alternating pattern of overdispersion and
underdispersion.
Best,
Colin and Dan
--
PhD Candidate in Government
Proctor
Harvard University
(617) 493-8402
Dear All,
Quick query: let's say I have a discrete, ordered variable (VAR) that takes on
values 1, 2, 3, or 4. I want to use R to create a dummy variable for those
observations that take on value 2. Up until now, I have been subsetting the
dataset into two pieces, defined by VAR==2 and ! VAR==2, then creating a new
variable (DATA$NEWVAR <- 1 for the first subset, DATA$NEWVAR <- 0 for the
second), and then putting the new datasets back together. There must be a more
efficient way to do this, and I was curious if anyone had any hints as to what
it was.
Thanks a lot!
Best,
Dan
--
PhD Candidate in Government
Proctor
Harvard University
(617) 493-8402
Hello,
I'm trying to impute some missing data in my dataset, but it's not
going too well.
There are two programs that can do imputation: Gary's program and
Schafer's program.
Gary's program
When I download Amelia, I get an icon my desktop. I click on it and
many more icons show up on the desktop. I don't know what file runs the
program. I clicked another Amelia icon, a window pops and immediately
disappears. I think Dan had sort of similar problem, but I can't even
run the program. HELP!!!
Schafer's program
I downloaded Schafer's Splus libraries on Science Center computers, but
I can't put them in Library folder because of user restriction. I have
Splus 2000, but Schafer's libraries are for Splus 4 and they are not
compatible. Kosuke speculates that there is a way to put libraries in a
local directory and make Splus recognize it. Any suggestions?
Thanks,
Chester
Dear All,
Quick question: when I log onto ICE4, I get the following message:
Last login: Wed Apr 9 18:46:56 2003 from net-39733.fas.harvard.edu
Could not chdir to home directory /home/d/h/dhopkins: No such file or directory
>
Colin gets the same result, although we are both on the ICE4 server. I was
curious 1) if others had this problem and 2) what I might do to fix it. I am
still able to access ICE3, although that's not where my VNC session is.
Many thanks, as always.
Best,
Dan
--
PhD Candidate in Government
Proctor
Harvard University
(617) 493-8402
This is a reminder that there is no section tomorrow. Since we had a
replication/comment assignment this week, Hw7 is due April 16. I'm around
this week, so if you want to stop by to talk about the paper project, send
me an e-mail.
Kosuke
FYI.
Kosuke
----------------------------
On Thursday, April 10th, 2003, from 6:00 AM to 7:00 AM, we will be
performing systems maintenance on the central FAS UNIX systems.
This maintenance may result in brief interruptions to all UNIX services,
including:
Login access
IMAP/POP/Webmail email access
Mail Delivery
Web and Portal services
DNS service
During this period, users should expect to lose any login sessions or
active processes they have on FAS and ICE servers.
We regret any inconvenience this may cause.
--
UNIX Systems Group
FAS Computer Services
Harvard University