I found "plogis" in the "base" library, and it appears to be giving me the
logit curve I'm looking for (much as "pnorm" generates the probit curve).
Can anyone confirm that this "logistic" is the same as the "logit" we're
looking for?
Thanks,
Ryan
------------------------------------------
Ryan T. Moore ~ Government & Social Policy
Ph.D. Candidate ~ Harvard University
How can we do it?
-p
-------------------------------------------------
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
-------------------------------------------------
This is a multi-part message in MIME format.
------=_NextPart_000_0153_01C2CF6C.D72AF710
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
What's the equation for the probit function, or where should I look to =
find it?
Thanks,
Ryan
------=_NextPart_000_0153_01C2CF6C.D72AF710
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 6.00.2600.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>What's the equation for the probit =
function, or=20
where should I look to find it?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>Thanks,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Ryan</FONT></DIV></BODY></HTML>
------=_NextPart_000_0153_01C2CF6C.D72AF710--
Since several students want to meet with me on Monday, I will hold an
extra office hour on Monday in my CBRSS office right after the lecture.
If you have any questions about R and other things, stop by then.
Kosuke
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Kosuke Imai
address:
Ph.D. candidate Center for Basic Research in the Social Sciences
Department of Government 34 Kirkland Street, Room 31
Harvard University Cambridge, MA 02138
phone: 617-496-3798, fax: 617-496-5149
URL: http://www.people.fas.harvard.edu/~kimai
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
One way to nice R within (X)emacs using ESS is to do the following:
1. add the following line, for example, in ``.aliases'' file in your home
directory:
alias R nice +10 R
2. Try ``source .aliases'' to make the change effective.
3. From now on, whenever you invoke R inside/outside of (X)emacs, it will
be niced.
Kosuke
--=====================_8505462==_.ALT
Content-Type: text/plain; charset="us-ascii"; format=flowed
Hey all,
I'm one of the people new to R, so I'm operating as if flying
blind in an F-16 with no idea what most of the controls do but I know how
flight works conceptually.
So, on problem 1b, I've got two vectors going, I've randomized one
of them (I think), and I am trying to compare the two, to see if the nth
unit in each vector (one randomized, one not -- again, I think, though if
you asked me to justify my code, I'd be shaky) is the same as the
corresponding in the other. What commands do I use to do something like this?
Thanks. Hope you're all having a good Friday night!
-Nathan
----------
Nathan A. Paxton, Ph.D. Student
Dept. of Government
Littauer Hall
Harvard University
Cambridge, Massachusetts 02138
<mailto:napaxton@fas.harvard.edu>napaxton(a)fas.harvard.edu
http://www.people.fas.harvard.edu/~napaxton/
========================================================================
The most courageous act is still to think for yourself. Aloud.
- Coco Chanel
When you have to stay eight years away from California, you live in a
perpetual state of homesickness.
- Ronald Reagan
========================================================================
--=====================_8505462==_.ALT
Content-Type: text/html; charset="us-ascii"
<html>
<x-tab> </x-tab>Hey
all,<br><br>
<x-tab> </x-tab>I'm one of
the people new to R, so I'm operating as if flying blind in an F-16 with
no idea what most of the controls do but I know how flight works
conceptually.<br><br>
<x-tab> </x-tab>So, on
problem 1b, I've got two vectors going, I've randomized one of them (I
think), and I am trying to compare the two, to see if the nth unit in
each vector (one randomized, one not -- again, I think, though if you
asked me to justify my code, I'd be shaky) is the same as the
corresponding in the other. What commands do I use to do something
like this?<br><br>
<x-tab> </x-tab>Thanks.
Hope you're all having a good Friday night!<br><br>
-Nathan<br>
<x-sigsep><p></x-sigsep>
----------<br>
Nathan A. Paxton, Ph.D. Student<br>
Dept. of Government<br>
Littauer Hall<br>
Harvard University<br>
Cambridge, Massachusetts 02138<br>
<a href="mailto:napaxton@fas.harvard.edu">napaxton(a)fas.harvard.edu</a><br>
<font color="#0000FF"><a href="http://www.people.fas.harvard.edu/~napaxton/" eudora="autourl">http://www.people.fas.harvard.edu/~napaxton/</a><br><br>
</font>========================================================================<br>
The most courageous act is still to think for yourself.
Aloud.<br>
<x-tab> </x-tab>- Coco
Chanel<br><br>
When you have to stay eight years away from California, you live in a
perpetual state of homesickness.<br>
<x-tab> </x-tab>- Ronald
Reagan<br>
========================================================================</html>
--=====================_8505462==_.ALT--
Dear Class,
I am a tad confused as I try to reproduce a logit function via R. The
equation given in class (Slide 21) is pi_i = 1 / (1 + exp(x_i*Beta)).
The graph on the same slide shows that pi varies from 0 to 1; it starts at 1
and then declines in an S shape, and for larger values of X seems to have an
asymptotic relationship with the x-axis.
Now, according to my interpretation of the equation, it should produces values
of pi that begin with .5 (when x_i = 0, the exponential term becomes 1
(anything to the zero power is 1) and thus the denominator is 2). From there,
I believe it should decrease, holding beta constant. But clearly, the graph
shows pi starting at 1. Am I thinking about this incorrectly, or is there
another explanation as to why the graph begins at 1 and not the .5 that I
think it should?
Best,
Dan
----- End forwarded message -----
I would like to have my R code spit info out while running (like where it
might be in the program, some values that it calculates to make sure things
make sense, and so on), but I want that output to go to a log (text) file.
I'm guessing there is an easy way to do this. Maybe:
R% cat (<STUFF I WANT TO PRINT OUT>) > <FILENAME>
Or something like that maybe? Thanks,
John.
Hi,
Is there a way to "clear" the memory allocated to objects and the like
within R (actually, I'm using S-Plus) without killing the R/S-Plus session?
Thanks,
John.
Just to finish responding to the "nice" question. To nice a process in
action, do the following:
Unix% renice <new value> <PID>
Where the new value cannot be lower than the existing one and PID is the
process ID. For example, say PID 0001 has a nice level of 5, then to nice it
to 10, you would:
Unix% renice +10 0001
And you would not be able to nice it to anything lower than 5. Hope this is
useful.
John.
