*RMS Short Course 2018*
Frank E. Harrell, Jr., Ph.D., Professor
Department of Biostatistics, Vanderbilt University School of Medicine
fharrell.com @f2harrell
*May 15-18, 2018* With Optional R Workshop May 14
9:00am - 4:00pm
Alumni Hall
Vanderbilt University
Nashville Tennessee USA
See
Fixed 'maxiter' in the help file. Thanks.
Please give the original source of that dataset.
That dataset is a tiny sample of GUSTO-I and not large enough to fit this
model very reliably.
A nomogram using the full dataset (not publicly available to my knowledge)
is already available in
In recent versions of rms on CRAN there was a non-downward compatible
change. To get latex output for summary, anova, and print on fit objects
you leave off file="" (because we're usually using knitr anyway) and use
options(prType='latex')
anova(f) # LaTeX output
You can use
A major new version of the rms package is now on CRAN. The most
user-visible changes are:
- interactive plotly graphic methods for model fits. The best example of
this is survplot for npsurv (Kaplan-Meier) estimates where the number of
risk pop up as you hover over the curves, and you can click
Hmisc 4.0-0 is now on CRAN. The package has undergone a massive update.
The most user-visable changes are;
- support for Rmarkdown html notebooks
- advanced html tables using the htmlTable package and summaryM function;
can copy and paste into word processors
- support for plotly interactive
This happens when you have not strat variables in the model.
--
Frank E Harrell Jr Professor and Chairman School of Medicine
Department of *Biostatistics* *Vanderbilt University*
On Thu, Jun 2, 2016 at 10:55 AM, Steve Lianoglou
I use this function to update my installed R packages:
updatePac <- function (checkBuilt = FALSE, ...)
update.packages(repos = "http://cran.rstudio.com;, instlib =
"/usr/local/lib/R/site-library",
checkBuilt = checkBuilt, ...)
When I type updatePac() I get:
Warning: package 'MASS' in
This does seem to be a good situation for ordinal regression. The R rms
package's orm function allows for thousands of categories in Y. But it
doesn't handle censoring.
This discussion would be better for stats.stackexchange.com
Frank
--
Rolf I believe \textsf is the correct font to use for the symbol R but
not necessarily for the names of R variables and functions. I'd like
more discussion about that.
Frank
--
Frank E Harrell Jr Professor and
Please note that Kirsten is cross-posting to stats.stackexchange.com
creating extra work for everyone.
--
Frank E Harrell Jr Professor and Chairman School of Medicine
Department of *Biostatistics*
New versions of Hmisc and rms are available on CRAN. Changes are listed
below.
The most significant change to Hmisc is the addition of the ffCompress
function that creates an optimal ff package object for large data frames
by computing the maximum number of bits used by each numeric or
It's implemented in the R rms package's lrm and orm functions.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Nagelkerke-Pseudo-R-squared-tp4710014p4710031.html
Sent from the R help mailing list archive
Thank you very much Terry. I'm still puzzled at why this worked a year
ago. What changed? I'd very much like to reverse the change by setting
an argument somewhere or manipulating the terms object.
I echo your sentiments about the general approach.
Frank
On 06/15/2015 09:05 AM, Therneau,
Terry - your example didn't demonstrate the problem because the variable
that interacted with strata (zed) was not a factor variable.
But I had stated the problem incorrectly. It's not that there are too
many strata terms; there are too many non-strata terms when the variable
interacting
For building design matrices for Cox proportional hazards models in the
cph function in the rms package I have always used this construct:
Terms - terms(formula, specials=c(strat, cluster, strata), data=data)
specials - attr(Terms, 'specials')
stra- specials$strat
Terms.ns - Terms
Several updates have been made to the Hmisc package:
Changes in version 3.16-0 (2015-04-25)
* html.contents.data.frame: corrected html space character to add
semicolon
* ggplot.summaryP: added size of points according to denominators
* colorFacet: new function
* labelPlotmath:
*RMS Short Course 2015*
Frank E. Harrell, Jr., Ph.D., Professor and Chair
Department of Biostatistics, Vanderbilt University School of Medicine
*March 3, 4, 5 6, 2015* With Optional R Workshop March 2
9:00am - 4:00pm
Student Life Center Board of Trust Room
Vanderbilt University
Nashville
Subject: Regression Modeling Strategies 4-Day Short Course March 2015
*RMS Short Course 2015*
Frank E. Harrell, Jr., Ph.D., Professor and Chair
Department of Biostatistics, Vanderbilt University School of Medicine
*March 3, 4, 5 6, 2015* With Optional R Workshop March 2
9:00am - 4:00pm
Student
. Thanks: Benjamin Tyner
* DESCRIPTION, NAMESPACE: multiple function changes to work in R-devel
See http://biostat.mc.vanderbilt.edu/Hmisc,
https://github.com/harrelfe/Hmisc for more information.
Frank Harrell
Vanderbilt University
___
R-packages
: start storing all classes created by survfit.formula
* logLik.Gls: added. Makes AIC(Gls object) work.
* NAMESPACE: several changes
See http://biostat.mc.vanderbilt.edu/rms and
https://github.com/harrelfe/rms for more information.
Frank Harrell
Vanderbilt University
When using cph in the rms package there is a function Mean that operates
on cph objects to produce an R function for computing the mean or
restricted mean life time.
Frank
__
R-help@r-project.org mailing list
Greg I just re-copied the latest subplot and its help file from
TeachingDemos to Hmisc for the next release. Thanks for pointing this out.
Frank
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the
Please try hard to create a simple reproducible example, then I'll work
on it.
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
I can't think of an example where R does not work better than SAS except
for a few cases of mixed effects regression models and for processing
enormous datasets when the R user does not want to learn about the
latest R tools for large datasets. I quit using SAS in 1991 (in favor
of S-Plus and
The Hmisc package has had a number of updates/fixes:
Changes in version 3.14-4 (2014-04-13)
* rcspline.eval: stop new logic for ignoring outer values when there
are many ties when there are also many ties on interior values. Added
new logic to use interior unique values of x when the
Recent changes include the following.
Changes in version 3.14-2 (2014-02-26)
* latex.default: improved logic using new function in Misc: latexBuild
* latex.default: fixed bug with ctable=TRUE with no caption by
removing default label
* latex.default: improved formatting for insert.top
Changes are listed below.
NOTE: The next release of rms will replace the survfit.formula function
with a new function named npsurv (nonparametric survival estimates) so
as to not conflict with the survival package. This is a
NON-DOWNWARD-COMPATIBLE change. To get Kaplan-Meier estimates in
The first submission of the new greport package is now on CTAN. This
package facilitates graphical reporting of clinical trials an is highly
tied to LaTeX, Hmisc, and lattice. It creates hyperlinks and pop-up
tooltips in the resulting pdf report. Tables are greatly de-emphasized,
but
My yearly Regression Modeling Strategies course is expanded to 4 days
this year to be able relax the pace a bit. Details are below.
Questions welcomed.
-
*RMS Short Course 2014*
Frank E. Harrell, Jr., Ph.D., Professor and Chair
.
Is there a way that i can see the progress of the read?
Or is there another way to read the file with less computing time?
I do not have access to SAS, the file was sent to me.
Let me know what you guys think
KR
Hans
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
Hmisc has had the following updates, with some of the new functions
demonstrated on http://biostat.mc.vanderbilt.edu/HmiscNew . Much of the
new work is motivated by the need to replace tables with graphics.
Changes in version
My yearly Regression Modeling Strategies course is expanded to 4 days
this year to be able relax the pace a bit. Details are below.
Questions welcomed.
-
*RMS Short Course 2014*
Frank E. Harrell, Jr., Ph.D., Professor and
This is an unstable process. I suggest using the bootstrap to get a
confidence interval for the rank of each correlation coefficient among all
non-diagonal correlations.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4
Thanks very much Rich and Duncan. latticeExtra's resizePanels function
was a perfect solution.
Frank
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
Is it possible to vary the size of the panels in lattice within one
page? The examine I have in mind is a 3 row by 2 column display where I
vary the y-axis scales in a dot plot and there are only a few levels on
the y-axis for one of the rows. I'd like to remove wasted space in that
row.
I would like to do this:
f - function(formula, data=NULL) {
gg - sqrt
model.frame(formula, data=data)
}
x - y - 1:10
f(y ~ gg(x))
Error in eval(expr, envir, enclos) : could not find function gg
Is there a simple way to get access to gg from within the model.frame
invocation inside f?
Thank you Bill, that worked perfectly.
Frank
__
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,
The rms package has had several updates in version 4.1-0:
* Fixed orm.fit to not create penalty matrix if not needed
(penalties are not yet implemented anyway)
* Added yscale argument to plot.Predict
* Added Wald test simulation to orm help file
* Added example in help file for
A significant update to the Hmisc package is now available on CRAN for
all platforms. Hmisc source is now on github at
https://github.com/harrelfe/Hmisc and the full change log may be found
at https://github.com/harrelfe/Hmisc/commits/master
The most important updates are additions of new
You did not follow the posting guide, did you use pure ascii email, and
used illegal characters in your source code. This caused extra work.
Once I cleaned up your characters and made the example self-contained,
the labels worked fine for me. Here's the cleaned-up code:
library(rms)
x1 -
Not certain about .por but this works with ordinary SPSS files:
require(Hmisc)
dat - spss.get(...) # gets variable labels, etc.
contents(dat)
html(contents(dat), ...)
The last command produces a hyperlinked data dictionary, e.g., for each
variable the number of levels is given and you click
This is dichotomania and is an inappropriate use of continuous
variables. Use an information-preserving approach such as rank
correlation. Also, this is not an R question.
Frank
---
Assume that I want to compare two methods, say skinfold measurement and
BMI, in how they classify
__
R-help@
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.
-
Frank Harrell
Department
require(rms)
?orm# ordinal regression model
For a case study see Handouts in
http://biostat.mc.vanderbilt.edu/CourseBios330
Since you have lost the original values, one part of the case study will
not apply: the use of Mean().
Frank
-
I have a dataset which has
Bill I found a workaround:
f - ff(formula, lab)
f - as.formula(gsub(`, , as.character(deparse(f
Thanks for your elegant solution.
Frank
--
Thanks Bill. The problem is one of the results of convertName might be
'Heading(Age in Years)*age' (this is for the
)
}
} else if (is.name(expr)) {
expr - as.name(convertName(expr))
}
expr
}
Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com
-Original Message-
From: [hidden email] [mailto:[hidden email]] On Behalf
Of Frank Harrell
Sent: Thursday, August 15, 2013 7:47
I really appreciate the excellent ideas from Bill Dunlap and Greg Snow.
Both suggestions almost work perfectly. Greg's recognizes expressions
such as sex=='female' but not ones such as age 21, age 21, a - b
0, and possibly other legal R expressions. Bill's idea is similar to
what Duncan
-Original Message-
From: [hidden email] [mailto:[hidden email]] On Behalf
Of William Dunlap
Sent: Thursday, August 15, 2013 6:03 PM
To: Frank Harrell; RHELP
Subject: Re: [R] regex challenge
Try this one
ff - function (expr)
{
if (is.call(expr) is.name(expr[[1
I would like to be able to use gsub or gsubfn to process a formula and
to translate the variables but to ignore expressions in the formula.
Supposing that the R formula has already been transformed into a
character string and that the transformation is to convert variable
names to upper case
(here on the x-axis) where the labels tell all. The last 2
lines in the code uses superpositioning for group but the col argument
affects the color of confidence bands; I didn't look into that further.
Frank Harrell
--
require(rms)
n = 30
group = factor(sample(c('a','b','c'), n, TRUE
Our Hmisc package summary.formula function and its latex methods can
make some fairly advanced tables. But the tables have to be regular.
For example, all rows of the tables are based on the same data frame.
I'm thinking that what is needed is a ggplot2-like set of functions for
building a
Duncan,
I had read your excellent tables package vignette at
http://cran.r-project.org/web/packages/tables/vignettes/tables.pdf when
it first came out. It is extremely impressive. I'm glad to be reminded
to give it another look.
Is there a way to make the special symbols n and 1 refer to
Thanks Rich and Jim and apologies for omitting the line
x - c(285, 43.75, 94, 150, 214, 375, 270, 350, 41.5, 210, 30, 37.6,
281, 101, 210)
But the fundamental problem remains that vertical spacing is not correct
unless I waste a lot of image space at the top.
Frank
--
Frank E Harrell Jr
[Sorry I can't quote past messages as I get mail on Nabble and have been
told I can't reply through Nabble].
Thanks Kennel for recommended a narrowing range for usr y-limits. That
does help quite a bit.
But I found a disappointing aspect of the graphics system: When you
change height= on
I can't get the points and a and b to render correctly (they are
squeezed into the axis) unless I make the height of the figure waste a
lot of space. I'd appreciate any ideas. The code is below. Thanks!
png('/tmp/z.png', width=480, height=100)
par(mar=c(3,.5,1,.5))
plot.new()
par(usr=c(-10,
The rms (Regression Modeling Strategies) package has undergone a
massive update. The entire list of updates is at the bottom of this
note. CRAN has the update for linux and will soon have it for Windows
and Mac - check http://cran.r-project.org/web/packages/rms/ for
availability. This rms
I have been confused about the appropriate e-mail address to use to make
announcements to r-help for major package update. In the past I've
submitted to r-packa...@lists.r-project.org without seeing the
announcement appear on r-help.
Thanks for any guidance.
Frank
--
Frank E Harrell Jr
Thank you Marc
Frank
Marc Schwartz-3 wrote
On Jul 10, 2013, at 1:29 PM, Frank Harrell lt;
f.harrell@
gt; wrote:
I have been confused about the appropriate e-mail address to use to make
announcements to r-help for major package update. In the past I've
submitted to
R-packages@.r
Mike,
Do something like:
require(rms)
dd - datadist(mydatarame); options(datadist='dd')
f - Rq(y ~ rcs(age,4)*sex, tau=.5) # use rq function in quantreg
summary(f) # inter-quartile-range differences in medians of y (b/c tau=.5)
plot(Predict(f, age, sex)) # show age effect on median as a
You are using an informal shrinkage method. It is much better to use
penalized maximum likelihood estimation, built in to lrm.
If you really want to go the informal route, compute the linear
predictor from your final estimates and use ols( ) to predict that from
the component variables
Unless you have detailed simulations to back up the performance of this
method I would avoid it. It violates several statistical principles.
Frank
Hari wrote
Hello R geeks,
Waiting for an reply.
Thanks,
Hari
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
). Is there an easy way to do this?
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/rmean-in-survfit-tp4667668p4667708.html
Sent from the R help mailing list archive at Nabble.com
-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Weighted-regression-in-rms-Hmisc-tp4667437p4667446.html
Sent
with this or suggest an R package ? I would
be very very grateful for help with this task !
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/how-to-do-an-external-validation-with-R-tp4667404p4667426.html
Sent from the R help
the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Tiff-plot-Resolution-issues
,
because its assumed discontinuities seldom exist in nature and most
relationships are not piecewise flat.
Frank
levanovd wrote
Or even simpler (no need to specify labels):
x-runif(100,0,100)
u - cut(x, breaks = c(0, 3, 4.5, 6, 8, Inf), labels = FALSE)
-
Frank Harrell
Department
commented, minimal, self-contained, reproducible code.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Stepwise-regression-for-multivariate-case-in-R-tp4665505p4665526.html
Sent from the R help mailing list
/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.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4
, summary.rms, validate))
ERROR: lazy loading failed for package ‘rms’
Any advice appreciated.
Frank
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/NAMESPACE-and-imports-tp4664718.html
Sent from the R help mailing
), importFrom
needs to know what functions you are importing [e.g. importFrom(Hmisc,
latex) importFrom(stats, anova)].
-Original Message-
From:
r-help-bounces@
[mailto:
r-help-bounces@
] On Behalf Of Frank Harrell
Sent: Freitag, 19. April 2013 16:26
To:
r-help@
Subject: [R
Now I see it. S3method() wants two arguments so I need to create multiple
S3method() statements for each generic.
Frank
Frank Harrell wrote
Right, I should have said import(Hmisc) instead of importFrom(Hmisc), but
that does not explain the error message.
Blaser Nello wrote
Not sure
What a terrific article. Thanks for sharing! The more we critically
examine how research is actually done the more frightened we become.
Frank
--
Frank E Harrell Jr Professor and Chairman School of Medicine
Department of Biostatistics Vanderbilt University
Does anyone know of R functions for doing composite quantile regression (Hou
and Yuan Ann Stat 36:1108, 2008)? The paper's authors do not talk about
software in their paper or on their web sites.
Thanks
Frank
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View
, there are thousands of such
combinations and R simply refuses to calculate them! (computer crashes) I
wonder if you can help me to write the code in R in order to find the
model wich include as many variables as it possible with significant
t-values?
cheers, Oleg
-
Frank Harrell
__
R-help@
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.
-
Frank Harrell
Department of Biostatistics, Vanderbilt
deleted]]
__
R-help@
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.
-
Frank
the accuracy, for example, for KNN,neural networks and J48, for
each one of 5-fold because when I apply CV to my classifier, I obtain the
mean accuracy of 5-fold but each accuracy/error of each fold is not
returned.
Any help is welcome and grateful. Thanks in advance!
Regards!!
-
Frank
m - length(bca)
bca - bca[c(m-1, m)]
}
list(np=np, bca=bca)
}
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Bootstrap-BCa-confidence-limits-with-your-own-resamples-tp4661045.html
Sent from the R
)
-- snip --
I hope this helps,
John
-Original Message-
From:
r-help-bounces@
[mailto:r-help-bounces@r-
project.org] On Behalf Of Frank Harrell
Sent: Tuesday, March 12, 2013 11:27 AM
To:
r-help@
Subject: [R] Bootstrap BCa confidence limits with your own
I'm not sure why you posted the original note. I quit using SAS in 1991 and
haven't needed it yet.
Frank
RogerJDeAngelis wrote
Sorry about the double post. But I keep getting 'post' rejections, so I
resubmitted about an hour later.
-
Frank Harrell
Department of Biostatistics
Achim this is perfect. I had not seen Formula before. Thanks for writing
it!
Frank
Achim Zeileis-4 wrote
On Fri, 1 Mar 2013, Frank Harrell wrote:
Thank you Bill. A temporary re-arrangement of the formula will allow me
to
do the usual subset= na.action= processing afterwards. Nice idea
, Frank Harrell lt;
f.harrell@
gt; wrote:
The lattice package uses special logic to allow for multiple
left-hand-side
variables in a formula, e.g. y1 + y2 ~ x. Is there an elegant way to
do
this outside of lattice? I'm trying to implement a data summarization
function that logically takes
) :
unimplemented type 'expression' in 'HashTableSetup'
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Expressions-in-lattice-conditional-variables-tp4660089.html
Sent from the R help mailing list archive
Whoops - these 2 lines should have been omitted from the program:
n - sprintf('%s (n%s=%g, n%s=%g)', v, nam[1],n[1], nam[2],n[2])
vn[var == v] - n
Frank Harrell wrote
I would like to have a lattice conditioning ( | var ) variable have
expression() as values because I want panel labels
]]
__
R-help@
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.
-
Frank Harrell
Department of Biostatistics
invocation of model.frame( ) causes R to try to do arithmetic addition to
create a single dependent variable.
Thanks
Frank
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Multiple-left-hand-side-variables
Thanks for your reply Gabor. That doesn't handle a mixture of factor and
numeric variables on the left hand side.
Frank
Gabor Grothendieck wrote
On Fri, Mar 1, 2013 at 7:16 PM, Frank Harrell lt;
f.harrell@
gt; wrote:
The lattice package uses special logic to allow for multiple
left-hand
@
[mailto:
r-help-bounces@
] On Behalf
Of Frank Harrell
Sent: Friday, March 01, 2013 4:17 PM
To:
r-help@
Subject: [R] Multiple left hand side variables in a formula
The lattice package uses special logic to allow for multiple
left-hand-side
variables in a formula, e.g. y1 + y2 ~ x
/xgsu00/
__
R-help@
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.
-
Frank
.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Plotting-survival-curves-after-multiple-imputation-tp4658552p4659505.html
Sent from the R help mailing list archive at Nabble.com
/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.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com
-guide.html
and provide commented, minimal, self-contained, reproducible code.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Relative-Risk-in-logistic-regression-tp4657040p4657163.html
Sent from the R help
*RMS Short Course 2013*
Frank E. Harrell, Jr., Ph.D., Professor and Chair
Department of Biostatistics, Vanderbilt University School of Medicine
*March 4, 5 6, 2013*
8:00am - 4:30pm
Student Life Center Board of Trust Room
Vanderbilt University
Nashville Tennessee USA
See
For lrm fits, predict(fit, type='mean') predicts the mean Y, not a
probability.
Frank
__
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
example and do the
calculations by hand.
Frank
blackscorpio81 wrote
Dear R users,
Please allow to me ask for your help.
I am currently using Frank Harrell Jr package rms to model ordinal
logistic regression with proportional odds. In order to assess model
predictive ability, C concordance
,
Please allow to me ask for your help.
I am currently using Frank Harrell Jr package rms to model ordinal
logistic regression with proportional odds. In order to assess model
predictive ability, C concordance index is displayed and equals to
0.963.
This is the code I used with the data attached
Chris,
I've fixed Surv in rms. The fix will be in the next release. For now you
can do source('http://biostat.mc.vanderbilt.edu/tmp/Surv.s') after issuing
require(rms).
Frank
Frank Harrell wrote
Chris,
Thanks for sending the specifics. It appears that I've let Surv in rms
fall behind
-contained, reproducible code.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/Does-psm-Surv-handle-interval2-data-tp4655472p4655475.html
Sent from the R help mailing list archive at Nabble.com
the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/suggestions-about-import-SAS
on individual variables.
Frank
-
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/suggestions-about-import-SAS-results-to-R-tp465p4654554.html
Sent from the R help mailing list archive at Nabble.com
1 - 100 of 452 matches
Mail list logo