survival analysis

3 messages · p.b.pynsent@bham.ac.uk, Michael A. Miller, Henrik Bengtsson

p.b.pynsent@bham.ac.uk

Fri, Nov 22, 2002 12:03 PM #

Frank,
You are quite right of course, the line:

rounded.fu <-ceiling(fuperiod)

does the trick.

As to why, this is what the J Bone Jnt Surg expect, although I agree  
that there is a loss of precision.

Any way thank you again.

On Thursday, November 21, 2002, at 03:32 pm, Frank E Harrell Jr wrote:

I'm not clear on why the actuarial method is still used, as it loses  
resolution by rounding follow-up times to the nearest month/year/week  
instead of using all the precision in the data.  Anyway if you  
rounding the follow-up times and used Kaplan-Meier, wouldn't you get  
actuarial estimates?

Frank Harrell

On Thu, 21 Nov 2002 13:54:00 +0000
Paul Pynsent <p.b.pynsent at bham.ac.uk> wrote:

Jun,
Thank you for this information. It is half of what I wanted as there
are no confidence limits calculated.
Survfit in the survival package gives very useful alternatives for
calculating the confidence limits including adjustments to the lower
limits, for data that is heavily censored (see Dorey and Korn as
referenced in the survfit documentation). On the other hand,
unfortunately survfit does not offer an actuarial method.
It looks as though I am going to have to write my own code for this.

On Wednesday, November 20, 2002, at 03:19  pm, Jun Yan wrote:

I am not sure if this is what you want. The function lifetab in  
package
KMsurv creates a cohort life table as described in Klein and
Moeschberger
(1997), Survival Analysis.

library(KMsurv)
?lifetab

Jun

On Wed, 20 Nov 2002, Paul Pynsent wrote:

Has anybody written an actuarial (life) survival procedure, this  
does
not appear to be an option in the survival package?
This approach is common in orthopaedic surgery to demonstrate the
survival of prostheses.
I need to apply the "modified" lower conf.int because of the  
censoring
over time.
I want a life table which I can then easily plot.
Many thanks

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 
.-
.-.-.-.-.-
r-help mailing list -- Read
http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._. 
_.
_._._._._


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
.-.-.-.-.-
r-help mailing list -- Read
http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._ 
._
._._._._

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 
.-.-.-.-.-
r-help mailing list -- Read  
http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:  
r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._. 
_._._._._


-- 
Frank E Harrell Jr              Prof. of Biostatistics & Statistics
Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
U. Virginia School of Medicine  http://hesweb1.med.virginia.edu/biostat

Dr. P. B. Pynsent,
Research and Teaching Centre,
Royal orthopaedic Hospital,
Birmingham, B31 2AP, U.K.

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Michael A. Miller

Fri, Nov 22, 2002 2:14 PM #

I have some data frames that contain logical values.  When I
write the frame to a file with write.table(df, filename,
row.names=F, sep=','), I end up with the logicals looking like
either ...,FALSE,... or ..., TRUE,... .  That space in front of
the TRUE is causing me problems when I read the frame again with
df <- read.table(filename,header=T,sep=',').  

Reading the data leaves me with this:

[1]  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE
[13]  TRUE FALSE FALSE  TRUE  TRUE  TRUE FALSE  TRUE  TRUE  TRUE  TRUE  TRUE
[25]  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE
Levels:   TRUE FALSE

[1]    NA    NA    NA    NA    NA    NA    NA    NA    NA    NA    NA    NA
[13]    NA FALSE FALSE    NA    NA    NA FALSE    NA    NA    NA    NA    NA
[25]    NA    NA    NA    NA    NA    NA

[1] " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE"
[10] " TRUE" " TRUE" " TRUE" " TRUE" "FALSE" "FALSE" " TRUE" " TRUE" " TRUE"
[19] "FALSE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE"
[28] " TRUE" " TRUE" " TRUE"

Can anyone suggest a way to read this column so that I can use it
as a logical without having to first do something like

df$valid <- as.logical(sub(' ','',as.character(df$valid)))

Mike

Michael A. Miller                               mmiller3 at iupui.edu
  Imaging Sciences, Department of Radiology, IU School of Medicine
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Henrik Bengtsson

Fri, Nov 22, 2002 6:38 PM #

A solution to your problem is to specify the argument colClasses:

df <- read.table(filename,header=T,sep=',', colClasses=colClasses)

where colClasses is a vector of character strings specifying the data
type of each column. For example:

data <- runif(30)
df <- data.frame(data=data, valid=(data > 0.5))
write.table(df, "tmp.dat", row.names=FALSE, sep=",")
df2 <- read.table("tmp.dat", header=TRUE, colClasses=c("double",
"logical"), sep=",")
print(as.logical(df2$valid)) 
#  [1]  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
TRUE
# [13]  TRUE  TRUE  TRUE  TRUE  TRUE FALSE FALSE  TRUE  TRUE  TRUE  TRUE
TRUE
# [25]  TRUE  TRUE  TRUE FALSE  TRUE FALSE

If you do not know the data type of a column in advance you can specify
it as NA (note not "NA"), e.g. colClasses=c(NA, "logical"). Specifying
the colClasses argument will also make read.table much faster. 

What is happening is that write.table() is first converting your data
frame to a matrix using as.matrix(). Try as.matrix(df) and there you see
why you get " TRUE" and not "TRUE". This is what is written to file.
When read.table() then reads the file it will not read it as a logical,
but as a factor variable, e.g.

sep=",")

[1]  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
TRUE
[13]  TRUE  TRUE  TRUE  TRUE  TRUE FALSE FALSE  TRUE  TRUE  TRUE  TRUE
TRUE
[25]  TRUE  TRUE  TRUE FALSE  TRUE FALSE
Levels:  TRUE FALSE

"logical"), sep=",")

[1]  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
TRUE
[13]  TRUE  TRUE  TRUE  TRUE  TRUE FALSE FALSE  TRUE  TRUE  TRUE  TRUE
TRUE
[25]  TRUE  TRUE  TRUE FALSE  TRUE FALSE

Note the last line "Levels:  TRUE FALSE" in the first case.

Best wishes

Henrik Bengtsson

Mathematical Statistics
Centre for Mathematical Sciences
Lund University

-----Original Message-----
From: owner-r-help at stat.math.ethz.ch 
[mailto:owner-r-help at stat.math.ethz.ch] On Behalf Of Michael A. Miller
Sent: den 23 november 2002 09:14
To: r-help at stat.math.ethz.ch
Subject: [R] read.table, write.table, logicals and spaces?

I have some data frames that contain logical values.  When I 
write the frame to a file with write.table(df, filename, 
row.names=F, sep=','), I end up with the logicals looking 
like either ...,FALSE,... or ..., TRUE,... .  That space in 
front of the TRUE is causing me problems when I read the 
frame again with df <- read.table(filename,header=T,sep=',').  

Reading the data leaves me with this:

df$valid

 [1]  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  TRUE  
TRUE  TRUE  TRUE [13]  TRUE FALSE FALSE  TRUE  TRUE  TRUE 
FALSE  TRUE  TRUE  TRUE  TRUE  TRUE [25]  TRUE  TRUE  TRUE  
TRUE  TRUE  TRUE
Levels:   TRUE FALSE

as.logical(df$valid)

 [1]    NA    NA    NA    NA    NA    NA    NA    NA    NA    
NA    NA    NA
[13]    NA FALSE FALSE    NA    NA    NA FALSE    NA    NA    
NA    NA    NA
[25]    NA    NA    NA    NA    NA    NA

as.character(df$valid)

 [1] " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" 
" TRUE" " TRUE" [10] " TRUE" " TRUE" " TRUE" " TRUE" "FALSE" 
"FALSE" " TRUE" " TRUE" " TRUE" [19] "FALSE" " TRUE" " TRUE" 
" TRUE" " TRUE" " TRUE" " TRUE" " TRUE" " TRUE" [28] " TRUE" 
" TRUE" " TRUE"

Can anyone suggest a way to read this column so that I can 
use it as a logical without having to first do something like

df$valid <- as.logical(sub(' ','',as.character(df$valid)))

Mike

-- 
Michael A. Miller                               mmiller3 at iupui.edu
  Imaging Sciences, Department of Radiology, IU School of Medicine
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
-.-.-.-.-.-.-.-.-
r-help mailing list -- Read 
http://www.ci.tuwien.ac.at/~hornik/R/R-> FAQ.html
Send "info", 
"help", or "[un]subscribe"
(in the 
"body", not the subject !)  To: 
r-help-request at stat.math.ethz.ch 
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
_._._._._._._._._

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._