You asked for a hint.
> library(MASS)
> x <- c(20.0, 23.9, 20.9, 23.8, 25.0, 24.0,
21.7, 23.8, 22.8, 23.1, 23.1, 23.5,
23.0, 23.0)
> fitdistr(x, "gamma")
shape rate
316.56387213.780766
(119.585534) ( 5.209952)
To do it with more general and e
Dear List,
I am not sure how should i optimize a log-likelihood numerically:
Here is a Text book example from Statistical Inference by George Casella, 2nd
Edition Casella and Berger, Roger L. Berger (2002, pp. 355, ex. 7.4 # 7.2.b):
data = x = c(20.0, 23.9, 20.9, 23.8, 25.0, 24.0, 21.7, 23.8, 22
Hi, all.
I need help on improving the efficiency of my R
simulations. Below is a function that simulates a
change point model. It first generates a sequence of
three dimensional ARMA(1,1) observations, then
calculates the one step ahead prediction errors, some
statistic is calcualted and compare
Hi,
I’m using the gls function to fit a model with autoregressive
residuals. Does anyone know how to do weighted least squares with
autoregressive residuals in gls?
Thanks,
Sincerely,
Xiaopeng
__
R-help@r-project.org mailing list
https://stat.ethz.
"Sudhindra Gadagkar" <[EMAIL PROTECTED]> wrote in
news:[EMAIL PROTECTED]:
> Hi, I can't believe I am unable to do this. I just installed R
> (never used it before, but want to learn it). At present, though.
> all I want to do is export a data file (specifically "crabs" in the
> MASS library) to
Sudhindra,
You should read the Posting Guide before asking for help:
http://www.R-project.org/posting-guide.html
It has greats hints on how to solve problems like this yourself, and you
should try them before posting.
It might suggest that this is a good thing to run in R if you need
first of all, which is your code to export the data in R? did you use
write.table?
secondly, it is totally a waste of time to export data from R for use
in spss. for whatever can be done by spss, r can handle better.
On Jan 26, 2008 9:38 PM, Sudhindra Gadagkar <[EMAIL PROTECTED]> wrote:
> Hi, I ca
Hi, I can't believe I am unable to do this. I just installed R (never used
it before, but want to learn it). At present, though. all I want to do is
export a data file (specifically "crabs" in the MASS library) to a text file
for use in SPSS. It allows me to peek into the dataset, using the comm
Here is a test I did with a data frame that had only 100 rows, but
the factors has 7000 and 4000 levels. I had to limit it to this
because with 7000 and 1 (which was your data size) I ran out of
physical memory on my system (1GB). You can see the amount of memory
that it was using which is d
On Jan 26, 2008 11:23 PM, Duncan Murdoch <[EMAIL PROTECTED]> wrote:
>
> On 26/01/2008 5:24 AM, Geoff Russell wrote:
> > Dear users,
> >
> > I'm trying to produce a 3d bar plot but the x and y dimensions have
> > categorical
> > data -- so I only want 3 points on each axis. So I try:
> >
> >
> > re
thanks, Jim.
My system has a RAM of 1 GB. I guess the computed matrix is close to the limit
of the memory and that's what caused the problem. I think I'll take Wensui's
suggestion and use a relational database system to handle the huge data.
> Date: Sat, 26 Jan 2008 20:40:51 -0500
> From: [
hi Gabor and Duncan,
I make a test and I find that the key of problem
is the setwd() command in my Rprofile.site
I don't understand this behaviour yet! :-(
Cleber
___
Experimente já e veja as novidades.
Thanks Henrik. I am using Rgui. Restart works. I was thinking about ways to
save the session, or prevent it from happening, in case this happens again.
I guess once it goes bad, nothing works except a hard restart. Is there an
auto save function in R?
On 1/26/08, Henrik Bengtsson <[EMAIL PROTECTED
I think with your data you will be computing a matrix that is 7049 x
11704. This will require about 700MB of memory. What size system do
you have (how much memory)? How big is the dataframe? (do 'str' and
report what it says). This will require a lot more resources and
given that you have about
hi Duncan,
Rename my Rprofile.site and then
this experiment works fine! because:
...
...
[Previously saved workspace restored]
but R --vanilla is not the same approach??
I try with 2 different files Rprofile.site,
the 2 Rprofile.site are many different
and the 2 has functions
with double-clicke
Well, I hesitate to discuss things like this on Rhelp, but I can tell
you about my experiences with them.
I have worked with them a bit. They have done R programming for my
company. For example, when odfWeave moved from heavy grepping to using
the xml package, they wrote the state-based xml parsin
I repeated your experiment:
> n <- 100
> x <- data.frame(A=sample(LETTERS,n,TRUE),
> B=sample(letters[1:4],n,TRUE),C=sample(LETTERS[1:4], n, TRUE), data=runif(n))
> system.time(x.agg <- aggregate(x$data, list(x$A, x$B, x$C), mean))
user system elapsed
1.824 0.212 2.038
Now I use m
How large is your dataframe? How much memory do you have on your
system? Are you paging? Here is a test I ran with a data frame with
1,000,000 entries and it seems to be fast:
> n <- 100
> x <- data.frame(A=sample(LETTERS,n,TRUE), B=sample(letters[1:4],n,TRUE),
+ C=sample(LETTERS[1:4],
On 26/01/2008, WCD <[EMAIL PROTECTED]> wrote:
>
> Yes, the switch function works fine.
> To store generated values in some variable I just use:
> switch(menu(c("Normal", "Uniform")), a<-rnorm(5), a<-runif(5))
>
> But why does not work this? (I know there is waste of variables, but in
> principle)
http://www.revolution-computing.com/revolution_company.html
seems to lead to something albeit vague.
Lots of apparently big names but no indication of what
added value the company might bring to an R user.
--- John Maindonald <[EMAIL PROTECTED]>
wrote:
> Does anyone know any more than is in t
On Jan 26, 2008 7:24 PM, Cleber Nogueira Borges <[EMAIL PROTECTED]> wrote:
> hi Gabor,
>
> Now, seems an light, because...
>
> C:\R\etc>assoc .Rdata
> .Rdata=RWorkspace
>
> C:\R\etc>ftype RWorkspace
> Tipo de arquivo 'RWorkspace' não encontrado ou nenhum comando open
> associado a ele.
> *File type
On 26/01/2008 5:31 PM, Cleber Nogueira Borges wrote:
> Gabor Grothendieck wrote:
>> Please clarify. What happens?
>
> ok!
> I start up the R, load my file by win menu: "Load Workspace",
> ls() show-me variables!
> Then, I save my workspace with other name (test2.rdata, for example) and
> close m
hi Gabor,
Now, seems an light, because...
C:\R\etc>assoc .Rdata
.Rdata=RWorkspace
C:\R\etc>ftype RWorkspace
Tipo de arquivo 'RWorkspace' não encontrado ou nenhum comando open
associado a ele.
*File type 'RWorkspace' not found or none command open associated for it*.
seems something with error
When and how does this happen? In Rgui or Rterm? How do you launch
R? Have you tried the obvious and restarted R? If you share the
startup messages, sessionInfo() and everything else you think could be
helpful, someone might be able to help you. There is a "vanilla"
options to start R, e.g.
R
How does the it compare if you read it into R and then do your
aggregate with sqldf:
library(sqldf)
# example using builtin data set CO2
CO2agg <- sqldf("select Plant, Type, Treatment, avg(conc) from CO2
group by Plant, Type, Treatment")
# or using your data:
Xagg <- sqldf("select Group, Age, T
huali,
if i were you, i will create a view on the MySql server to aggregate
the data first and then use R to pull the data through this created
view. This is not only applicable to R but also a general guideline in
similar situation.
Per my understanding and experience, R is able to do data manipul
Sorry, indeed I forgot to put some of the factors in the code. Here it
is again, now updated:
calcurve <- cbind(1:2e4, 1:2e4, rep(100, length=2e4));
caldist <- function(cage, error, sdev=2, times=5, By=1)
{
calcurve <- calcurve[which((calcurve[,2]+calcurve[,3]) >=
cage-(times*error)),];
Hi,
My R just froze. I can't get it to do anything. It gives "Error: band value"
message to everything I type. Does anyone know if R has a safe mode that I
could check for errors and perform diagnostics? I am using R 2.6.1 on
Windows XP.
> ls()
Error: bad value
> search()
Error: bad value
> ?ls
Er
Try entering this at the Windows console:
assoc .Rdata
assuming the answer is:
.Rdata=RWorkspace
try this:
ftype RWorkspace
That will show you what command is executed when you double
click on an Rdata file. Perhaps you need to change that.
On Jan 26, 2008 6:40 PM, Cleber Nogueira Borges <[E
Hi, netters,
First of all, thanks a lot for all the prompt replies to my earlier question
about "merging" data frames in R.
Actually that's an equivalence to the "join" clause in mysql.
Now I have another question. Suppose I have a data frame X with lots of
columns/variables:
Name, Age,Group,
Does anyone know any more than is in the following press release
about REvolution Computing and their commercialization of R?
http://www.intel.com/capital/news/releases/080122.htm
"Intel Capital, the global investment arm of Intel Corporation, today
announced that it has invested in the Series
hi Gabor,
I try it:
in *MS-DOS*:
C:\R\etc>Rgui --vanilla
In *R*:
a=1
b=2
c=3
save.image("c:\\R\\etc\\test.rdata")
q('no')
then I "double-clicked' in 'test.rdata' (in c:\R\etc folder)
The R start without my variables: a,b,c
If I make again: R -- vanilla, and then load file, the variables are the
You could try
Rgui --vanilla
so that it does not run the scripts you have set up.
On Jan 26, 2008 5:31 PM, Cleber Nogueira Borges <[EMAIL PROTECTED]> wrote:
> Gabor Grothendieck wrote:
> > Please clarify. What happens?
>
> ok!
> I start up the R, load my file by win menu: "Load Workspace",
> ls
Your function has 4 parameters and you are only calling with two. The
first statement:
caldist <- function(cage=Cage, error=Error, sdev=Sdev, times=Times, By=By)
{
theta <- seq(min(calcurve[,1]), max(calcurve[,1]), by=By);
use "By" which is not defined.
On Jan 26, 2008 4:29 PM, Maarten Blaau
"Dupouey Jean-Luc" <[EMAIL PROTECTED]> wrote in
news:[EMAIL PROTECTED]:
On my system wilcox.test() exists, whereas wilcox_test() did not.
Help.search("wilcox_test") does not produce any clues. If you are using
a test from another package, you should specify which package the test
comes from.
Gabor Grothendieck wrote:
> Please clarify. What happens?
ok!
I start up the R, load my file by win menu: "Load Workspace",
ls() show-me variables!
Then, I save my workspace with other name (test2.rdata, for example) and
close my E session.
When I start up R by double-click in my new RData file
Please clarify. What happens?
On Jan 26, 2008 5:02 PM, Cleber Nogueira Borges <[EMAIL PROTECTED]> wrote:
>
> hi,
> I tried this approach, but don't work too!
> (Duncan Murdoch make this suggestion)
>
> Cleber
>
> > OK. The system only knows about one R version.
> >
> > What about the answer to m
Have a look at rbind.fill() in the reshape library.
This will return a data.frame not a matrix but you can
always convert it to a matrix.
--- zhihuali <[EMAIL PROTECTED]> wrote:
>
> Hi netters,
> Suppose I have two data frames X and Y. X has three
> colnames A, B and C. Y has three colnames A,
hi,
I tried this approach, but don't work too!
(Duncan Murdoch make this suggestion)
Cleber
> OK. The system only knows about one R version.
>
> What about the answer to my original second question.
> If you start up R and then load the .Rdata file (which you
> previously indicated works) and t
Lately R has been behaving strange on my Linux (Ubuntu 7.10) machine,
with occasional segfaults. Today something else and reproducible
happened:
If I type the code below (meant for calibrating data), I get the error
message that "the C stack usage is too close to the limit".
calcurve <- cbi
RSQLite 0.6-7 has been uploaded to CRAN and should hit a mirror near
you in the next few days.
This version changes the behavior of the dbGetQuery method to make it
more consistent with dbSendQuery. Specifically:
1. dbGetQuery now closes a complete result set as dbSendQuery does.
2. If there is
OK. The system only knows about one R version.
What about the answer to my original second question.
If you start up R and then load the .Rdata file (which you
previously indicated works) and then
save your workspace using a different name from the original
Rdata file and then double click on tha
hello Gabor,
the result in combobox is:
R version 2.6.1 (2007-11-26) - C:\R
I not have Perl installed! I remember that I deleted it!
> Just double click on Rversions.hta from Windows Explorer.
> Alternately place it anywhere in your path and then enter
> its name in the Windows console without
On 26/01/2008, David Winsemius <[EMAIL PROTECTED]> wrote:
> "Henrique Dallazuanna" <[EMAIL PROTECTED]> wrote in
> news:[EMAIL PROTECTED]:
>
> > Try this:
> > merge(x, y, all=T)
>
> That will give 's in the non-matching rows, which may be what the
> OP wanted. If, on the otherhand, only the rows wit
"Henrique Dallazuanna" <[EMAIL PROTECTED]> wrote in
news:[EMAIL PROTECTED]:
> Try this:
> merge(x, y, all=T)
That will give 's in the non-matching rows, which may be what the
OP wanted. If, on the otherhand, only the rows with matches were
desired, the usage might be:
A.B <- merge(X, Y, by =
Dear R-forumites,
I want to apply a Wilcoxon test on subsets of the data frame mydata,
splitted using the myindice variable. When I send :
wilcoxtest <- by(mydata, mydata$myindice, function(x)
{wilcox_test(x$value~x$fact)})
I get :
Error in eval(expr, envir, enclos) : object "x" not
Hi, Huali,
there is another way to do so
library(sqldf)
wanted <- sqldf("select * from X inner join Y on X.A = Y.A and X.B = Y.A")
2008/1/26 zhihuali <[EMAIL PROTECTED]>:
>
> Hi netters,
> Suppose I have two data frames X and Y. X has three colnames A, B and C. Y
> has three colnames A,B and D.
Try this:
merge(x, y, all=T)
On 26/01/2008, zhihuali <[EMAIL PROTECTED]> wrote:
>
> Hi netters,
> Suppose I have two data frames X and Y. X has three colnames A, B and C. Y
> has three colnames A,B and D.
>
> I want to combine them into one matrix, joining the rows having the same A
> and B va
dear liviu,
first of all, i appreciate your insight.
i've been used r on windows for years and might use it on unix. i
understand r is similar across OSs. However, per my understanding,
still there is nontrivial difference between running on unix and
running on windows.
have a nice weekend.
On Jan
Henrique Dallazuanna wrote:
> Try this:
>
> tmpfile <- tempfile()
> download.file('http://www.stat.ucla.edu/projects/datasets/risk_project.dta',
> tmpfile)
> risk2 <- read.dta(tmpfile)
> unlink(tmpfile)
The stata.get function in the Hmisc package will automatically fetch
from http://... and prov
Hi netters,
Suppose I have two data frames X and Y. X has three colnames A, B and C. Y has
three colnames A,B and D.
I want to combine them into one matrix, joining the rows having the same A and
B values (X$A==Y$A and X$B = Y$B). So the resulting dataframe has four
variables/columns: A,B,C an
Yes, the switch function works fine.
To store generated values in some variable I just use:
switch(menu(c("Normal", "Uniform")), a<-rnorm(5), a<-runif(5))
But why does not work this? (I know there is waste of variables, but in
principle):
b<-numeric(10)
switch(menu(c("Normal", "Uniform")),
Hello,
On 1/26/08, Wensui Liu <[EMAIL PROTECTED]> wrote:
> dear all,
> is there a tutorial specifically teaching how to use R under unix? it
> seems most of them are about using r under window.
> thank you so much!
What exactly are you interested in: installing or using R? If it is
the latter, R
Reparameterize to
f(x, y) = sqrt(x*x) * exp(y*y)
On Jan 26, 2008 1:11 PM, roger chan <[EMAIL PROTECTED]> wrote:
>
> Hi, I have a scalar valued function with several variables. One of the
> variables is restricted to be non-negative. For example,
> f(x,y)=sqrt(x)*exp(y), then x should be non-negat
Try this:
tmpfile <- tempfile()
download.file('http://www.stat.ucla.edu/projects/datasets/risk_project.dta',
tmpfile)
risk2 <- read.dta(tmpfile)
unlink(tmpfile)
On 26/01/2008, Michael Kubovy <[EMAIL PROTECTED]> wrote:
> Dear R-helpers,
>
> I would like to have my students read into R an online S
When you double click it does it start up the same version of R
as when you manually start R?
If you manually start R, load it, save it under another name and
then exit R and double click the new name does that work?
On Jan 26, 2008 1:01 PM, Cleber Nogueira Borges <[EMAIL PROTECTED]> wrote:
> hel
Hi, I have a scalar valued function with several variables. One of the
variables is restricted to be non-negative. For example,
f(x,y)=sqrt(x)*exp(y), then x should be non-negative. I need the
gradient and hessian for some vector (0,y), i.e., I need the gradient and
hessian at the boudary of par
dear all,
is there a tutorial specifically teaching how to use R under unix? it
seems most of them are about using r under window.
thank you so much!
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the pos
On 26/01/2008 1:01 PM, Cleber Nogueira Borges wrote:
> hello all,
>
> when I start up the R and I execute o follow code:
>
> > ls()
> character(0)
> > x=123
> > assign("test_x", x, envir = .GlobalEnv )
> > ls()
> [1] "test_x" "x"
> > setwd('C:\\R\\etc')
> > save.image('TEST.RData')
>
Try this also:
noquote(tapply(x$Abundance, list(x$Plot, x$Species), paste))
On 25/01/2008, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>
> Hello.
>
> Does anyone know of an existing function that takes a list in the form of:
>
> Plot1 Species1Abundance1
> Plot1 Species2Abundance2
> Plot
hello all,
when I start up the R and I execute o follow code:
> ls()
character(0)
> x=123
> assign("test_x", x, envir = .GlobalEnv )
> ls()
[1] "test_x" "x"
> setwd('C:\\R\\etc')
> save.image('TEST.RData')
> q('no')
I have two different behaviours:
(a) - when I start up R again by "
Try like this:
switch(menu(c("Normal", "Poisson")), rnorm(5), rpois(5, 3))
On 25/01/2008, WCD <[EMAIL PROTECTED]> wrote:
>
> Hello, I would like to make my R script more interactive. Well, I want the
> script to work like this:
>
> I run the script from R Console and it offers me some options (p
Hi,
Not going into the reason why would you like to do such a thing anyway, I
would suggest:
1. Creating an R file that would contain all the functionality coded in
functions.
2. The only executed code would be a function that calls the menu() and
executes the appropriate functions.
Sources o
Many thanks to all. These methods all provide the needed info:
library(help= lattice)
packageDescription("lattice ", fields="Built")
packageDescription("lattice")$Built
The first method produces nore complete information, should that also be
interesting.
Charles Annis, P.E.
[EMAIL PROTECTED
hits=-2.6 tests=BAYES_00
X-USF-Spam-Flag: NO
Dear Geoff,
scatterplot3d was never intended to draw a plot for categorical data,
but you can tweak it as follows:
- Note that the help page says (stolen from ?par) "The values of x and
y give the (approximate) number of tickmarks on the x and y a
On 26/01/2008 11:44 AM, Charles Annis, P.E. wrote:
> Greetings, R-ians:
>
>
>
> I would like to know which version or R was used to create a given package.
> I think I remember seeing that topic discussed recently but cannot find it
> among my notes. Can anyone tell me how to determine which v
These also works should you need to capture it:
packageDescription("lattice")$Built
On Jan 26, 2008 12:04 PM, Gabor Csardi <[EMAIL PROTECTED]> wrote:
> > library(help=lattice)
> [...]
> Built: R 2.6.0; i486-pc-linux-gnu; 2008-01-23 13:52:49; unix
> [...]
>
> G.
>
>
> On Sat, Jan 26, 2008
> library(help=lattice)
[...]
Built: R 2.6.0; i486-pc-linux-gnu; 2008-01-23 13:52:49; unix
[...]
G.
On Sat, Jan 26, 2008 at 11:44:53AM -0500, Charles Annis, P.E. wrote:
> Greetings, R-ians:
>
>
>
> I would like to know which version or R was used to create a given package.
> I think I
Greetings, R-ians:
I would like to know which version or R was used to create a given package.
I think I remember seeing that topic discussed recently but cannot find it
among my notes. Can anyone tell me how to determine which version of R
created a package?
Thanks.
Charles Annis, P.E
Dear R-help,
In nlme package, anova () can be used to test the difference between
two coefficients as shown on page 225
of "Mixed Effects Models in S and S-Plus":
anova(fm2BW.lme, L = c("Time:Diet2" = 1, "Time:Diet3" = -1))
Now my question is instead of test the difference between two
coefficien
Try this (adding 0 converts the logicals to numeric):
rowsum((X > 0)+0, group)
On Jan 26, 2008 9:13 AM, Anders Bjørgesæter <[EMAIL PROTECTED]> wrote:
> Hello
>
> Is there an analogous function to "rowsum" that count the numbers
> according to a given vector (preferably larger than a given value)
On 26/01/2008 9:17 AM, Dennis Fisher wrote:
> Colleagues,
>
> I am preparing scripts that will be used by others on both Windows and
> Linux/OSX platforms. The scripts call an existing Fortran
> application. The user may have any of a a variety of Fortran
> compilers - my goal is to determ
Is this what you are looking for in doing computations on the columns
of a dataframe?
> x <-
> data.frame(group=sample(1:2,10,TRUE),A=sample(1:4,10,TRUE),B=sample(1:5,10,TRUE),
+ C=sample(1:3,10,TRUE))
> x
group A B C
1 1 2 4 1
2 2 4 1 2
3 1 3 2 1
4 2 3 2 2
5 2 4 2
Colleagues,
I am preparing scripts that will be used by others on both Windows and
Linux/OSX platforms. The scripts call an existing Fortran
application. The user may have any of a a variety of Fortran
compilers - my goal is to determine whether or not the test command
returns "no input
Hello
Is there an analogous function to ”rowsum” that count the numbers
according to a given vector (preferably larger than a given value)
instead of summing them?
E.g. rowsum(x, group)
X is the dataframe, A B C
5 0 0
Hallo,
Ihre E-Mailadresse "[EMAIL PROTECTED]" wurde in den Newsletterverteiler
hinzugefügt.
URL: http://www.fueralles.de/cgi-bin/nsl_pro/newsletter.cgi?id=steinfeuerfred
Mit freundlichen Grüssen Ihr Newsletter-Team
__
R-help@r-project.org mailing l
Hi
I am using ggplot2 at the moment and I must say it is definitely better
then ggplot - good work.
My problem is that I am using facet_grid() in the following way:
> p <- ggplot(ssq, aes(x=year, y=-log(ssq)))
> p + geom_point() + facet_grid(me*gi~cs*rz)
and it works nicely, except that I wo
Dear Alyaa,
you've now sent this message in many different forms to the
R-help mailing list.
The fact that you did not get a response may very well be
related to the way you asked your question,
but also that it probably is not an easy question to answer.
But sending your e-mail repeatedly to *s
Hi,
The following code, from Angelo Canty article on line "Resampling Methods in
R: the boot Package, 2002", works fine for Angelo Canty using R 2.6.0 on
Windows XP.
It also works for me using R 1.2.1 and S-PLUS 2000 on Windows XP after
installing the S-PLUS bootstrap library
Dear R-helpers,
I would like to have my students read into R an online Stata dataset:
'http://www.stat.ucla.edu/projects/datasets/risk_project.dta'
I was able to read it into R after downloading it and converting it
with StatTransfer (http://www.stattransfer.com/).
Here is what happens when I
Dear users,
I'm trying to produce a 3d bar plot but the x and y dimensions have categorical
data -- so I only want 3 points on each axis. So I try:
require(scatterplot3d)
mymat<-data.frame(
x=c(1,1,1,2,2,2,3,3,3),
y=c(1,2,3,1,2,3,1,2,3),
z=c(1,2,3,4,5,6,7,8,9))
scatterplot3d(myma
Hi, The following code, from Angelo Canty article on line
"Resampling Methods in R: the boot Package, 2002", works fine for Angelo Canty
using R 2.6.0 on Windows XP.
It also works for me using R 1.2.1 and S-PLUS 2000 on Windows XP after
installing the S-PLUS boots
Hi, The following code, from Angelo Canty article on line "Resampling
Methods in R: the boot Package, 2002", works fine for Angelo Canty using R
2.6.0 on Windows XP.
It also works for me using R 1.2.1 and S-PLUS 2000 on Windows XP after
installing the S-PLUS bootstrap library,
Is there a way in GAMM to increase the amount of smoothing (something like
min.sp in GAM)?
This is a large-data problem with binomial response. I can get something
like the desired effect by
using very few knots, but this seems like a kludge.
thanks!
Scott Olsson
[[alternative HTML ver
84 matches
Mail list logo
