3 Jan
2003
3 Jan
'03
11:09 p.m.
Since we're obviously not getting a solution for the "lattice problem" and
everybody has better things to do, here are some tips for setting up lattice:
To start:
library(nlme)
lset(theme = col.whitebg())
Try out:
demo(lattice)
Basic functions and applications (I think there's more but this is what I've
figured out so far - use help() for details):
histogram(~x | y, data = blah) --> here, "y" needs to be coded as a
character
variable. Then you get y numbers of graphs labeled properly. If you don't care
about the labels, "y" can be numeric.
xyplot(y ~ x | z, data = blah) --> same thing for "z" as above
cloud(y ~ x * z, data = blah) --> this creates a 3D plot. The variables don't
have to be characterized.
also potentially interesting:
levelplot()
densityplot()
"panel" and "group" are useful parameters, but I don't think I
fully understand
the details...
Still haven't figured out how to put in the abline...
Best,
Phillip
ps - feel free to buy me a beer :)
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
4 Jan
4 Jan
12:03 a.m.
Thanks, Phil. That's very helpful.
While playing around with plot() and xyplot(), have you come across any
instructions for plotting two different lines on the same graph? If I
remember correctly, GK have something like that in their paper -- where
one line is solid and the other is dotted.
- Nirmala
On Fri, 3 Jan 2003, Phillip Y. Lipscy wrote:
Since we're obviously not getting a solution for
the "lattice problem" and
everybody has better things to do, here are some tips for setting up lattice:
To start:
library(nlme)
lset(theme = col.whitebg())
Try out:
demo(lattice)
Basic functions and applications (I think there's more but this is what I've
figured out so far - use help() for details):
histogram(~x | y, data = blah) --> here, "y" needs to be coded as a
character
variable. Then you get y numbers of graphs labeled properly. If you don't care
about the labels, "y" can be numeric.
xyplot(y ~ x | z, data = blah) --> same thing for "z" as above
cloud(y ~ x * z, data = blah) --> this creates a 3D plot. The variables don't
have to be characterized.
also potentially interesting:
levelplot()
densityplot()
"panel" and "group" are useful parameters, but I don't think I
fully understand
the details...
Still haven't figured out how to put in the abline...
Best,
Phillip
ps - feel free to buy me a beer :)
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
12:15 a.m.
I'm not sure if this is what you're looking for, but this will give you a whole
bunch of lines:
groups = blah, panel=panel.superpose, type="l"
-p
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
Quoting Nirmala Ravishankar <ravishan(a)fas.harvard.edu>du>:
Thanks, Phil. That's very helpful.
While playing around with plot() and xyplot(), have you come across any
instructions for plotting two different lines on the same graph? If I
remember correctly, GK have something like that in their paper -- where
one line is solid and the other is dotted.
- Nirmala
On Fri, 3 Jan 2003, Phillip Y. Lipscy wrote:
Since we're obviously not getting a solution
for the "lattice problem"
and
everybody has better things to do, here are some
tips for setting up
lattice:
To start:
library(nlme)
lset(theme = col.whitebg())
Try out:
demo(lattice)
Basic functions and applications (I think there's more but this is what
I've
figured out so far - use help() for details):
histogram(~x | y, data = blah) --> here, "y" needs to be coded as a
character
variable. Then you get y numbers of graphs
labeled properly. If you don't
care
about the labels, "y" can be numeric.
xyplot(y ~ x | z, data = blah) --> same thing for "z" as above
cloud(y ~ x * z, data = blah) --> this creates a 3D plot. The variables
don't
have to be characterized.
also potentially interesting:
levelplot()
densityplot()
"panel" and "group" are useful parameters, but I don't think I
fully
understand
the details...
Still haven't figured out how to put in the abline...
Best,
Phillip
ps - feel free to buy me a beer :)
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
12:29 a.m.
Actually, I was trying to use two different dataframe to generate two
simple plot.default() lines in the same graph and this seems to work:
plot(a$year, a$psi, type = "b")
lines(b$year, b$psi, type = "l")
- Nirmala
On Sat, 4 Jan 2003, Phillip Y. Lipscy wrote:
I'm not sure if this is what you're looking
for, but this will give you a whole
bunch of lines:
groups = blah, panel=panel.superpose, type="l"
-p
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
Quoting Nirmala Ravishankar <ravishan(a)fas.harvard.edu>du>:
Thanks, Phil. That's very helpful.
While playing around with plot() and xyplot(), have you come across any
instructions for plotting two different lines on the same graph? If I
remember correctly, GK have something like that in their paper -- where
one line is solid and the other is dotted.
- Nirmala
On Fri, 3 Jan 2003, Phillip Y. Lipscy wrote:
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
Since we're obviously not getting a solution
for the "lattice problem"
and
everybody has better things to do, here are some
tips for setting up
lattice:
To start:
library(nlme)
lset(theme = col.whitebg())
Try out:
demo(lattice)
Basic functions and applications (I think there's more but this is what
I've
figured out so far - use help() for details):
histogram(~x | y, data = blah) --> here, "y" needs to be coded as a
character
variable. Then you get y numbers of graphs
labeled properly. If you don't
care
about the labels, "y" can be numeric.
xyplot(y ~ x | z, data = blah) --> same thing for "z" as above
cloud(y ~ x * z, data = blah) --> this creates a 3D plot. The variables
don't
have to be characterized.
also potentially interesting:
levelplot()
densityplot()
"panel" and "group" are useful parameters, but I don't think I
fully
understand
the details...
Still haven't figured out how to put in the abline...
Best,
Phillip
ps - feel free to buy me a beer :)
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
10:22 p.m.
New subject: Strange Error
Dear All,
So, every so often I get this strange error while doing a relatively
simple operation. It happens occassionally while doing something quite
routine and goes away if I quit R, erase all my objects, load them again,
and start over.
ca <- redist[redist$state %in% c(71),]
Error
in match(x, table, nomatch = 0) : unused argument(s) (nomatch ...)
Does this happen to anyone else? Do you know why R does this, or better
still, how to make it stop?
Thanks,
Nirmala
5 Jan
5 Jan
9:40 p.m.
New subject: Strange Error
Nirmala Ravishankar writes:
Dear All,
So, every so often I get this strange error while doing a relatively
simple operation. It happens occassionally while doing something quite
routine and goes away if I quit R, erase all my objects, load them again,
and start over.
Hmmm.
What does traceback() say? Have you learned how to use the browser?
What do the first 10 objects in redist$state looks like? Could this be
a factor, numeric, character thing? Does the snippet:
redist$state %in% c(71)
producing something sensible by itself? You want it to produce a
vector of TRUE/FALSE's for use in subsetting.
Dave
ca <- redist[redist$state %in% c(71),]
Error in match(x, table, nomatch = 0) : unused argument(s) (nomatch ...)
Does this happen to anyone else? Do you know why R does this, or better
still, how to make it stop?
Thanks,
Nirmala
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
--
David Kane
Lecturer in Government
617-563-0122
dkane(a)latte.harvard.edu
4 Jan
4 Jan
12:16 a.m.
To get "abline" to work, I think we have to figure out "panel." I
have
been trying to make sense of help(panel.grid), but with little success.
- Nirmala
On Fri, 3 Jan 2003, Phillip Y. Lipscy wrote:
Since we're obviously not getting a solution for
the "lattice problem" and
everybody has better things to do, here are some tips for setting up lattice:
To start:
library(nlme)
lset(theme = col.whitebg())
Try out:
demo(lattice)
Basic functions and applications (I think there's more but this is what I've
figured out so far - use help() for details):
histogram(~x | y, data = blah) --> here, "y" needs to be coded as a
character
variable. Then you get y numbers of graphs labeled properly. If you don't care
about the labels, "y" can be numeric.
xyplot(y ~ x | z, data = blah) --> same thing for "z" as above
cloud(y ~ x * z, data = blah) --> this creates a 3D plot. The variables don't
have to be characterized.
also potentially interesting:
levelplot()
densityplot()
"panel" and "group" are useful parameters, but I don't think I
fully understand
the details...
Still haven't figured out how to put in the abline...
Best,
Phillip
ps - feel free to buy me a beer :)
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
5 Jan
5 Jan
9:37 p.m.
Phillip Y. Lipscy writes:
Since we're obviously not getting a solution for
the "lattice problem" and
everybody has better things to do, here are some tips for setting up lattice:
Thanks for forwarding this to the list. Although there are no points
for class participation in GOV 1000, public-spiritedness on the e-mail
list is encouraged and rewarded. (This is not directed at Philip, who
is clearly public-spirited enough that he would send out this e-mail
without any extra-points bribery.)
By the way, it is not clear if "everybody has better things to do". Of
course, if GOV 1000 were ever taught again the way that it has been
taught this year (perish the thought!), many things would be "easier"
for the students. As an example, the early problem sets would provide
more hints about helpful R commands. However, a major purpose of the
class is to teach you how to figure out things for yourself. Alas, the
only way to learn how to do this is to practice it. So, even in a
world in which I gave out The World's Perfect Guide to Lattice, I
would still ask you questions that went beyond the guide, indeed ones
that go beyond my own knowledge.
It is painful and frustrating and infuriating to bang your head
against the wall and struggle for a solution that, after you find it,
takes only a few lines of code to create. It is also a critical
skill to have if you want to push the boundaries of empirtical
research in political science. And so we practice it . . .
Dave
To start:
library(nlme)
lset(theme = col.whitebg())
Try out:
demo(lattice)
Basic functions and applications (I think there's more but this is what I've
figured out so far - use help() for details):
histogram(~x | y, data = blah) --> here, "y" needs to be coded as a
character
variable. Then you get y numbers of graphs labeled properly. If you don't care
about the labels, "y" can be numeric.
xyplot(y ~ x | z, data = blah) --> same thing for "z" as above
cloud(y ~ x * z, data = blah) --> this creates a 3D plot. The variables don't
have to be characterized.
also potentially interesting:
levelplot()
densityplot()
"panel" and "group" are useful parameters, but I don't think I
fully understand
the details...
Still haven't figured out how to put in the abline...
Best,
Phillip
ps - feel free to buy me a beer :)
-------------------------------------------------
Phillip Y. Lipscy
Perkins Hall Room #129
35 Oxford Street
Cambridge, MA 02138
(617)493-4893
lipscy(a)fas.harvard.edu
First Year Student, Ph.D. Program
Harvard University, FAS, Department of Government
-------------------------------------------------
_______________________________________________
gov1000-list mailing list
gov1000-list(a)fas.harvard.edu
http://www.fas.harvard.edu/mailman/listinfo/gov1000-list
--
David Kane
Lecturer in Government
617-563-0122
dkane(a)latte.harvard.edu
7804
days inactive
7806
days old
gov1000-list@lists.fas.harvard.edu
7 comments
3 participants
participants (3)
-
Dave Kane -
Nirmala Ravishankar -
Phillip Y. Lipscy