[R] rms package: residual df for penalised ols

Dear R-help, In the rms package, when using the ols function with a penalty, the df.residual appears to always be n-1 (with n being the sample size). That seems strange to me, but I don't have much knowledge in this area. Here's an example: library(rms) set.seed(1) n <- 50 d <- data.frame(x1

Re: [R] glmnet: converting coefficients back to original scale

Thanks for your reply Mehmet. I've found that the problem was that I didn't scale the lambda value. My original example did not follow the instruction not to give a single lambda value, but that in itself wasn't the problem. Example shown below. library(glmnet) library(MASS) set.seed(1) n <- 20

[R] glmnet: converting coefficients back to original scale

Dear R-help, I'm having trouble understanding how glmnet converts its coefficient estimates back to the original scale. Here's an example with ridge regression: library(glmnet) set.seed(1) n <- 20 # sample size d <- data.

[R] rms package: error with Glm

nyone can give. Mark Seeto # library(rms) set.seed(1) n <- 100 # sample size beta0 <- 3.7 beta1 <- 1.5 beta2 <- 0.9 beta3 <- 0.5 rate.x1 <- 2 mean.x2 <- 1 sd.x2 <- 2 nu <- 1.3 d <- dat

[R] oblique.scores argument for fa function in psych package

% fa2$Structure)[1:3, ] # not the same as fa2$scores[1:3, ] ### Have I misunderstood something? Thanks, Mark Seeto __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE d

Re: [R] Gls function in rms package

ation=corARMA(p=2), method="ML") anova(gls.intercept, gls.rcs4) Thanks, Mark Seeto National Acoustic Laboratories, Australia > Appears to be a definite bug, probably caused by having more than one > correlation parameter. I hope to have this fixed within 3 days. > Frank &

[R] Gls function in rms package

represent a variable with a spline using rcs. I'm using version 3-5.0 of rms in R 2.15.0. Thanks for any help you can give. Mark Seeto __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting gu

[R] Warnings when plotting after x11() in R 2.14.2

also happens with ggplot2 plots. I'm using RGui on Windows 7. It did not happen with R 2.13.1. It's not a major problem, because the plots still appear to be produced correctly, but if anyone can tell me how to fix it, I'd appreciate it. Thanks, Mark Seeto __

[R] Column name containing "-"

5 y 1 00 0 2 11 1 names(d)[2] <- "a.-5" d x a.-5 y 1 00 0 2 11 1 Why does the "a.-5" column name change to "a..5" when another column is added? Thanks, Mark Seeto __ R-help@r-project.org mailing list h

[R] Replying on Nabble

to the mailing list. If I reply to the list, there's an option to email the post to someone, but I don't know the person's email address. Thanks, Mark Seeto -- View this message in context: http://r.789695.n4.nabble.com/Replying-on-Nabble-tp3691214p3691214.html Sent from the

Re: [R] a question on plotting nonlinear regression

Mark Seeto wrote: > > > garciap wrote: >> >> Hi to all the people, >> >> I'm having a trouble when trying to plot a quadratic function. I have the >> code: >> >> regression<-nls(Survival~beta1+beta2*PI+beta3*PI^2, data=cubs, >>

Re: [R] a question on plotting nonlinear regression

garciap wrote: > > Hi to all the people, > > I'm having a trouble when trying to plot a quadratic function. I have the > code: > > regression<-nls(Survival~beta1+beta2*PI+beta3*PI^2, data=cubs, > start=list(beta1 = 1, beta2 = 1, beta3 = 1)) > plot(Survival~PI,data=cubs, ylab="Survival", xlab="P

Re: [R] R graphs differ from exported one

forget to run dev.off(), as in that example. > > But we don't have the > >   commented, minimal, self-contained, reproducible code. > > the posting guide and the footer of every R message asks for. > > On Sun, 12 Jun 2011, Mark Seeto wrote: > >> >> Raptorist

Re: [R] R graphs differ from exported one

Raptorista wrote: > > Now, the graph that appears is very nice: indeed it has a title, two > axes with their labels and all the rest; > but when I give commands > > postscript(file="plot.eps", onefile=FALSE) > qqnorm (col) > > to save the graph to a file "plot.eps" to include it in a TeX, the f

Re: [R] predict with model (rms package)

Thanks for your reply, Frank. I've noticed that the x.knots object doesn't actually have to be the vector of knots. Just having x.knots <- 0 or even x.knots <- "a" will allow predict to work. Mark Seeto Frank Harrell wrote: > > This is a consequence of pred

[R] predict with model (rms package)

is simply defined as a vector like c(-1, 0, 1) (i.e. not using quantile). Is this the intended behaviour? The requirement that x.knots be in the workspace seems strange, given that the knot locations are stored in ols1$Design$parms. Thanks for any help you can give. Mark Seeto National Acoustic L

Re: [R] Model formula for ols function (rms package)

want the restricted interaction operator: y ~ rcs(x1, 3) + rcs(x2, 3) + rcs(x1, 3) %ia% rcs(x2, 3). For the second example use pol(x,2) or something like pol(x1,2) + pol(x2,2) + pol(x1, 2) %ia% pol(x2, 2) If you have to create new variables for R formulas you're usually doing something wrong

[R] Model formula for ols function (rms package)

term Is there a way to do these things without first creating new variables in the data frame? Thanks, Mark Seeto National Acoustic Laboratories __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posti

[R] rms: getting adjusted R^2 from ols object

7;t see where to go from there. Thanks, Mark Seeto __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.

Re: [R] Help with Amelia

Amit Patel-7 wrote: > > Hi > I have used the amelia command from the Amelia R package. this gives me a > number > of imputed datasets. > > This may be a silly question, but i am not a statistician, but I am not > sure how > to combine these results to obtain the imputed dataset to usse for f

Re: [R] Urgent Help with R calculation correlation coefficient

Try excluding the first column. cor(gse20437[, 2:4]) chintan85 wrote: > > > Tab delimited file looks like this > > Id v1 v2v3 > df 56 9045 > gh 87 9878 > ty 897867 > > I used this code > > > [code] > > gse20437 <- read.csv("C:/Users//Desktop/data/GSE20437_matrix

[R] When to use bootstrap confidence intervals?

n.samples = 1000, df = 3, boot.reps = 1000) $t [1] 0.027 0.014 0.959 $bca [1] 0.054 0.047 0.899 I don't understand the warning message, but for these examples, the ordinary t interval appears to be better than the bootstrap BCA interval. I would really appreciate any recommendations anyone

Re: [R] Multiple imputation, especially in rms/Hmisc packages

same? Thanks in advance; I really appreciate any help you can give. Mark Frank Harrell wrote: On Mon, 9 Aug 2010, Mark Seeto wrote: Hello, I have a general question about combining imputations as well as a question specific to the rms and Hmisc packages. The situation is multiple regressio

Re: [R] Multiple imputation, especially in rms/Hmisc packages

Thank you for your reply Frank. I am not familiar with the contrast test, but I'll see what I can find out about it. Mark Frank Harrell wrote: On Mon, 9 Aug 2010, Mark Seeto wrote: Hello, I have a general question about combining imputations as well as a question specific to the rm

[R] Multiple imputation, especially in rms/Hmisc packages

result in the ERROR line. I would be most grateful if anyone could explain this to me. Thanks, Mark -- Mark Seeto Statistician National Acoustic Laboratories <http://www.nal.gov.au/> A Division of Australian Hearing __ R-help@r-project.org mail

[R] Model validation and penalization with rms package

ive with this R^2 optimism? It can be decreased by taking a bigger penalty, but then the corrected R^2 is reduced. Also, a penalty of 9 gives a corrected slope of about 1.17 (corrected slope of 1 is achieved with a penalty of about 1 or 2). Thanks for any help/advice you ca

Re: [R] Specifying formula inside a function

Bill and Erik, thank you very much for your help. In addition to solving my problem, both solutions contain other good things I didn't know about. Regards, Mark Seeto On Thu, Jun 10, 2010 at 2:44 PM, Erik Iverson wrote: > Hello, > >> How does one specify a formula to lm inside

[R] Specifying formula inside a function

utput will show "formula = y ~ x3 + x4" instead of "formula = paste..."? Thanks for any help you can give. Regards, Mark -- Mark Seeto Statistician National Acoustic Laboratories A Division of Australian Hearing __ R-help@r-project.org mai

Re: [R] ols function in rms package

> On 06/08/2010 05:29 AM, Mark Seeto wrote: >> >>> On 06/06/2010 10:49 PM, Mark Seeto wrote: >>>> Hello, >>>> >>>> I have a couple of questions about the ols function in Frank Harrell's >>>> rms >>>> package. >

Re: [R] ols function in rms package

> On 06/06/2010 10:49 PM, Mark Seeto wrote: >> Hello, >> >> I have a couple of questions about the ols function in Frank Harrell's >> rms >> package. >> >> Is there any way to specify variables by their column number in the data >> f

[R] ols function in rms package

Error in terms.formula(formula) : '.' in formula and no 'data' argument Thanks for any help you can give. Regards, Mark -- Mark Seeto Statistician National Acoustic Laboratories <http://www.nal.gov.au/> A Division of Australian Hearing 126 Greville Street Chatswoo

Re: [R] Overfitting/Calibration plots (Statistics question)

Thank you very much for your help, Prof. Harrell. I was making the bad mistake of judging the appearance of the calibration plots without actually calculating the regression line. I was misjudging slopes of 0.8 or 0.9 as being slopes greater than 1. Kind regards, Mark Seeto > Mark, >

[R] Overfitting/Calibration plots (Statistics question)

xing up the axes). I would be most appreciative if anyone could explain where I'm going wrong. Thanks for any help you can provide. Mark Seeto __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.

Re: [R] Double inequality with plotmath

Thank you very much Duncan. I see I also had the mistake of a "c" instead of "expression" in my question. Mark On Fri, Jan 22, 2010 at 7:59 AM, Duncan Murdoch wrote: > On 21/01/2010 3:50 PM, Mark Seeto wrote: >> >> Hello, >> >> I'm fairly n

[R] Double inequality with plotmath

Hello, I'm fairly new to R and I can't work out how to produce a double inequality like (LaTeX) $0 \leq x \leq 1$ in the legend of a graph. If I try > legend(50, 0.1, legend = c(expression(0 <= x <= 1), c(2 <= x <= 3)), pch = > c(1,1), col = c(2, 3)) then I get an error message "unexpected '<='