This is a multi-part message in MIME format.
------=_NextPart_000_0027_01C2E4A6.01B61750
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
in the PDF given on the homework, in the first fraction, is the square =
root sign in the denominator over v and pi only or over v, pi,and sigma?
T
------=_NextPart_000_0027_01C2E4A6.01B61750
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2719.2200" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>in the PDF given on the homework, in =
the first=20
fraction, is the square root sign in the denominator over v and pi only =
or over=20
v, pi,and sigma?<BR><BR>T</FONT></DIV></BODY></HTML>
------=_NextPart_000_0027_01C2E4A6.01B61750--
This is a multi-part message in MIME format.
------=_NextPart_000_0032_01C2E4A6.EF87BE70
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Oh, and is pi 3.14 or some parameter?
T
------=_NextPart_000_0032_01C2E4A6.EF87BE70
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2719.2200" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Oh, and is pi 3.14 or some =
parameter?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2><BR>T</FONT></DIV></BODY></HTML>
------=_NextPart_000_0032_01C2E4A6.EF87BE70--
Can someone clarify what the concept of degree of freedom means? Does it
refer to the sample size? Also, in regression analysis, if you add
interaction terms, does that aggravate the degree of freedom problem? If so,
why and how? If not, why not?
yongwook
-----------------------------
Yongwook Ryu
PhD Student
Department of Government
Harvard University
Tel:617-493-3397
Email: yryu(a)fas.harvard.edu
-----------------------------
1. Is there a specific way to specify the size of the bounding box for graphics
files? I've been saving my R output as .eps files, but after converting to
.pdf, I still get error messages related to "no bounding box" when running pdflatex.
2. I've also been trying to convert .jpg files I have into .pdf/.eps so I can
put them into a LaTeX document, but I consistently get the "no bounding box"
error. Is there a good way to convert such files so they can be imported into a
LaTeX document?
3. Was there something unique about the "\bibliographystyle{pa}" thing we were
using in Gov 1000? It doesn't seem to work on the FAS server (I'm using
"plainnat" instead, but the citations look weird). Is there an easy way to
install it?
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
-------------------------------------------------
Hello!
We are endeavoring to load a space-delimited dataset where NA values are also
blank spaces into R. The dataset is available in two separate forms, "SPSS"
(.por) or raw (.dat).
1. When we try to use the read.spss command in the foreign library, it
complains that:
> tab.1 <- read.spss("~/SocCap/datatest1", use.value.labels=F)
Error in read.spss("~/SocCap/datatest1", use.value.labels = F) :
Error reading portable-file dictionary.
In addition: Warning message:
Duplicate label for value 0.5 for variable WWWTIME.
2. When we try to use the scan command, it creates an ENORMOUS vector:
> tab.1 <- scan("~/SocCap/usmisc2000-soccap.dat", what="complex", fill=T)
Read 7789818 items
> summary(tab.1)
Length Mode
7789818 character
The central problem is this: the SPSS file is tab delimited and has lots
of "junk"; the .dat file is space delimited but also has NAs denoted by spaces.
How can the computer tell spaces that separate the data from spaces that are
missing data?
Any thoughts or suggestions would be much appreciated.
Best,
Dan and Colin
This is a MIME-formatted message. If you see this text it means that your
E-mail software does not support MIME-formatted messages.
--=_latte.harvard.edu-23738-1046895067-0001-2
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Dear Graduate Students,
Listed below are two short-listed candidates for the quantitative political
methodology junior faculty position in the Government Department. Please
attend their job presentations if you can, and please contact me at
496-1512 or kolodney(a)hdc.harvard.edu if you would like to sign up to attend
their candidate-student breakfasts, or if you have any other
questions. The breakfasts are scheduled from 8:30-9:45 at the Inn at
Harvard prior to their job presentations.
In addition, I have circulated an email from Gary King inviting your views
on the candidate selection process should you choose to do so.
Please le me know if you have any questions.
Sincerely,
Ruth
Jonathan Wand (Cornell University degree candidate--June 2003 and a CBRSS
Research Fellow)
Date of Job Visit: Tuesday, March 11
Title of Job Talk: "The Party is in the PACs: How special interest groups
fund partisan electoral competition"
Location: TBA
Time: 10:00-11:30
Kevin M. Quinn (Ph.D. Washington University, St. Louis; currently Assistant
Professor, Department of Political Science, University of Washington)
Date of Job Visit, Friday, March 14
Title of Job Talk: "Re-Examining the Impact of Public Opinion on U.S.
Supreme Court Decision-Making"
Location: CBRSS, 34 Kirkland Street
Time: 10:30-12
--=_latte.harvard.edu-23738-1046895067-0001-2
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<html>
<br><br>
Dear Graduate Students,<br>
Listed below are two short-listed candidates for the quantitative
political methodology junior faculty position in the Government
Department. Please attend their job presentations if you can, and
please contact me at 496-1512 or kolodney(a)hdc.harvard.edu if you would
like to sign up to attend their candidate-student breakfasts, or if you
have any other questions. The breakfasts are scheduled from
8:30-9:45 at the Inn at Harvard prior to their job presentations.
<br>
In addition, I have circulated an email from Gary King inviting your
views on the candidate selection process should you choose to do
so.<br>
Please le me know if you have any questions.<br>
Sincerely,<br>
Ruth<br><br>
<b>Jonathan Wand (Cornell University degree candidate--June 2003 and a
CBRSS Research Fellow)<br>
</b>Date of Job Visit: Tuesday, March 11<br>
Title of Job Talk: "The Party is in the PACs: How special interest
groups fund partisan electoral competition"<br>
Location: TBA<br>
Time: 10:00-11:30<br><br>
<b>Kevin M. Quinn (Ph.D. Washington University, St. Louis; currently
Assistant Professor, Department of Political Science, University of
Washington)<br>
</b>Date of Job Visit, Friday, March 14<br>
Title of Job Talk: "Re-Examining the Impact of Public Opinion on
U.S. Supreme Court Decision-Making"<br>
Location: CBRSS, 34 Kirkland Street<br>
Time: 10:30-12<br><br>
<br>
</html>
--=_latte.harvard.edu-23738-1046895067-0001-2--
For 2b, if we use the product term of the negative binomial distribution to
estimate the likelihood rather than the loglikelihood, am I right to assume that
we should get the same result for the maximum point? i.e. we use loglikelihood
instead of likelihood to get rid of the product term, but our results should not
change?
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
-------------------------------------------------
The joint distribution of (Z,W) are given in the question. What you want
is the joint distribution (Z-W, Z+W). Use the theorem to get this dist:
find a matrix A s.t. A(Z,W)=(Z-W, Z+W). If you find the covariance of this
distribution to be zero, you can assert they are independent.
Kosuke
> Hi Kosuke,
>
> I think understand the idea of doing 1(a) analytically -- you want a proof
> using matrix algebra, right? So Z and W are i.i.d. on N(mu, sigma^2). What is
> the analytical (mathematical) definition of "independence"? (Seems like I'll
> need that to work my way from Z and W i.i.d. to Z+W and Z-W i.i.d.)
>
> Hope this question makes sense!
>
This is a multi-part message in MIME format.
------=_NextPart_000_0158_01C2E2A8.BDDD5F60
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
has anybody else been kicked off the system? My vnc died and now fas =
won't let me back on even when i type nice + 10 Xvnc etc.
T
------=_NextPart_000_0158_01C2E2A8.BDDD5F60
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2719.2200" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>has anybody else been kicked off the =
system? =20
My vnc died and now fas won't let me back on even when i type nice + 10 =
Xvnc=20
etc.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>T</FONT></DIV></BODY></HTML>
------=_NextPart_000_0158_01C2E2A8.BDDD5F60--