Re: [R] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Paul Miller]

Thanks for the clarification Dr. Therneau. Until I learn more about this I can 
at least remember that "plain" is bad.

Thanks,

Paul

--- On Mon, 4/16/12, Terry Therneau  wrote:

> From: Terry Therneau 
> Subject: Re: Kaplan Meier analysis: 95% CI wider in R than in SAS
> To: r-help@r-project.org, "Paul Miller" 
> Received: Monday, April 16, 2012, 8:30 AM
> On 04/14/2012 05:00 AM, r-help-requ...@r-project.org
> wrote:
> > Am replicating in R an analysis I did earlier using
> SAS. See this as a test of whether I'm ready to start using
> R in my day-to-day work.
> > ?
> > Just finished replicating a Kaplan Meier analysis.
> Everything seems to work out fine except for one thing. The
> 95% CI around my estimate for the median is substantially
> larger in R than in SAS. For example, in SAS I have a median
> of 3.29 with a 95% CI of [1.15, 5.29]. In R, I get a median
> of 3.29 with a 95% CI of [1.35,?13.35].
> > ?
> > Can anyone tell me why I get this difference?
> > 
> 
> The confidence interval for the median is based on the
> confidence intervals for the curves.  There are several
> methods for computing confidence intervals for the curves:
> plain, log, log-log, or logit scale.  There are
> opinions on which is best, and it is a close race: except
> for the first of these.  The type "plain" intervals are
> awful, it's like putting me in one lane of a championship
> 100 meter dash.
> 
> Until about version 9 the only option in SAS was "plain",
> then for a time it was still the default.  By 9.2 they
> finally went to loglog.
> 
> Terry Therneau
>

__
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] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Enrico Colosimo]

But it is very very unlikely to have a time with survival
probability of 0. in a real data set. It would be
necessary huge data set. A Monte Carlo simulation could
put a little more light in this issue?

2012/4/16 Frank Harrell 

> Just generate some data where the estimated survival probability is 0.
> at
> a certain time.  The log-log transformation blows up.
> Frank
>
> Enrico Colosimo wrote
> >
> > What are the significant problems of the log-log transformations?
> > Any papers published about it?
> > Enrico.
> >
> >
> > 2012/4/14 Frank Harrell 
> >
> >> I used log-log in my book too until Terry Therneau alerted me to the
> >> significant problems this creates.  In the 2nd edition it will use log
> >> S(t).
> >> Frank
> >>
> >> Paul Miller wrote
> >> >
> >> > Hello Drs. Colosimo and Harrell,
> >> >
> >> > Thank you for your replies to my question. From Dr. Colosimo, I was
> >> able
> >> > to determine that the SAS results can be replicated by adding the
> >> > option conf.type="log-log" to my code as in :
> >> >
> >> > survobj <- survfit(survfrm, conf.type="log-log", data=Survival)
> >> >
> >> > Originally, it looked like the SAS results could be replicated using
> >> > conf.type="plain". Applying this option to my actual data revealed
> that
> >> > this was not the case, however.
> >> >
> >> >>From Dr. Harrell, I learned that using conf.type="log-log" may not be
> >> such
> >> a good idea. Interestingly though, I've seen at least one instance where
> >> experts in the R community use this option in their book. The book is
> >> about
> >> 10 years old. So maybe opinion about the use of this option has shifted
> >> since then.
> >> >
> >> > Thanks,
> >> >
> >> > Paul
> >> >
> >> >   [[alternative HTML version 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 Harrell
> >> Department of Biostatistics, Vanderbilt University
> >> --
> >> View this message in context:
> >>
> http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4557695.html
> >> Sent from the R help mailing list archive at Nabble.com.
> >>
> >> __
> >> 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.
> >>
> >
> >   [[alternative HTML version 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 Harrell
> Department of Biostatistics, Vanderbilt University
> --
> View this message in context:
> http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4561432.html
> Sent from the R help mailing list archive at Nabble.com.
>
> __
> 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.
>

[[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 http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Terry Therneau]

On 04/14/2012 05:00 AM, r-help-requ...@r-project.org wrote:

Am replicating in R an analysis I did earlier using SAS. See this as a test of 
whether I'm ready to start using R in my day-to-day work.
?
Just finished replicating a Kaplan Meier analysis. Everything seems to work out 
fine except for one thing. The 95% CI around my estimate for the median is 
substantially larger in R than in SAS. For example, in SAS I have a median of 
3.29 with a 95% CI of [1.15, 5.29]. In R, I get a median of 3.29 with a 95% CI 
of [1.35,?13.35].
?
Can anyone tell me why I get this difference?



The confidence interval for the median is based on the confidence 
intervals for the curves.  There are several methods for computing 
confidence intervals for the curves: plain, log, log-log, or logit 
scale.  There are opinions on which is best, and it is a close race: 
except for the first of these.  The type "plain" intervals are awful, 
it's like putting me in one lane of a championship 100 meter dash.


Until about version 9 the only option in SAS was "plain", then for a 
time it was still the default.  By 9.2 they finally went to loglog.


Terry Therneau

__
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] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Frank Harrell]

Just generate some data where the estimated survival probability is 0. at
a certain time.  The log-log transformation blows up.
Frank

Enrico Colosimo wrote
> 
> What are the significant problems of the log-log transformations?
> Any papers published about it?
> Enrico.
> 
> 
> 2012/4/14 Frank Harrell 
> 
>> I used log-log in my book too until Terry Therneau alerted me to the
>> significant problems this creates.  In the 2nd edition it will use log
>> S(t).
>> Frank
>>
>> Paul Miller wrote
>> >
>> > Hello Drs. Colosimo and Harrell,
>> >
>> > Thank you for your replies to my question. From Dr. Colosimo, I was
>> able
>> > to determine that the SAS results can be replicated by adding the
>> > option conf.type="log-log" to my code as in :
>> >
>> > survobj <- survfit(survfrm, conf.type="log-log", data=Survival)
>> >
>> > Originally, it looked like the SAS results could be replicated using
>> > conf.type="plain". Applying this option to my actual data revealed that
>> > this was not the case, however.
>> >
>> >>From Dr. Harrell, I learned that using conf.type="log-log" may not be
>> such
>> a good idea. Interestingly though, I've seen at least one instance where
>> experts in the R community use this option in their book. The book is
>> about
>> 10 years old. So maybe opinion about the use of this option has shifted
>> since then.
>> >
>> > Thanks,
>> >
>> > Paul
>> >
>> >   [[alternative HTML version 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 Harrell
>> Department of Biostatistics, Vanderbilt University
>> --
>> View this message in context:
>> http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4557695.html
>> Sent from the R help mailing list archive at Nabble.com.
>>
>> __
>> 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.
>>
> 
>   [[alternative HTML version 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 Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context: 
http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4561432.html
Sent from the R help mailing list archive at Nabble.com.

__
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] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Enrico Colosimo]

What are the significant problems of the log-log transformations?
Any papers published about it?
Enrico.


2012/4/14 Frank Harrell 

> I used log-log in my book too until Terry Therneau alerted me to the
> significant problems this creates.  In the 2nd edition it will use log
> S(t).
> Frank
>
> Paul Miller wrote
> >
> > Hello Drs. Colosimo and Harrell,
> >
> > Thank you for your replies to my question. From Dr. Colosimo, I was able
> > to determine that the SAS results can be replicated by adding the
> > option conf.type="log-log" to my code as in :
> >
> > survobj <- survfit(survfrm, conf.type="log-log", data=Survival)
> >
> > Originally, it looked like the SAS results could be replicated using
> > conf.type="plain". Applying this option to my actual data revealed that
> > this was not the case, however.
> >
> >>From Dr. Harrell, I learned that using conf.type="log-log" may not be
> such
> a good idea. Interestingly though, I've seen at least one instance where
> experts in the R community use this option in their book. The book is about
> 10 years old. So maybe opinion about the use of this option has shifted
> since then.
> >
> > Thanks,
> >
> > Paul
> >
> >   [[alternative HTML version 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 Harrell
> Department of Biostatistics, Vanderbilt University
> --
> View this message in context:
> http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4557695.html
> Sent from the R help mailing list archive at Nabble.com.
>
> __
> 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.
>

[[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 http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Frank Harrell]

I used log-log in my book too until Terry Therneau alerted me to the
significant problems this creates.  In the 2nd edition it will use log S(t).
Frank

Paul Miller wrote
> 
> Hello Drs. Colosimo and Harrell,
>  
> Thank you for your replies to my question. From Dr. Colosimo, I was able
> to determine that the SAS results can be replicated by adding the
> option conf.type="log-log" to my code as in :
>  
> survobj <- survfit(survfrm, conf.type="log-log", data=Survival)
>  
> Originally, it looked like the SAS results could be replicated using
> conf.type="plain". Applying this option to my actual data revealed that
> this was not the case, however. 
>  
>>From Dr. Harrell, I learned that using conf.type="log-log" may not be such
a good idea. Interestingly though, I've seen at least one instance where
experts in the R community use this option in their book. The book is about
10 years old. So maybe opinion about the use of this option has shifted
since then. 
>  
> Thanks,
>  
> Paul  
>  
>   [[alternative HTML version 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 Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context: 
http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4557695.html
Sent from the R help mailing list archive at Nabble.com.

__
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] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Paul Miller]

Hello Drs. Colosimo and Harrell,
 
Thank you for your replies to my question. From Dr. Colosimo, I was able to 
determine that the SAS results can be replicated by adding the 
option conf.type="log-log" to my code as in :
 
survobj <- survfit(survfrm, conf.type="log-log", data=Survival)
 
Originally, it looked like the SAS results could be replicated using 
conf.type="plain". Applying this option to my actual data revealed that this 
was not the case, however. 
 
>From Dr. Harrell, I learned that using conf.type="log-log" may not be such a 
>good idea. Interestingly though, I've seen at least one instance where experts 
>in the R community use this option in their book. The book is about 10 years 
>old. So maybe opinion about the use of this option has shifted since then. 
 
Thanks,
 
Paul  
 
[[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 http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Frank Harrell]

Make sure you use the log S(t) basis on both systems (and avoid log-log S(t)
basis as this results in instability in the front part of the survival
curve).
Frank

Paul Miller wrote
> 
> Hi Enrico,
> 
> Not sure how SAS builds the CI but I can look into it. The SAS
> documentation does have a section on computational formulas at:
> 
> http://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_lifetest_a000259.htm
> 
> Although I can't provide my dataset, I can provide the data and code
> below. This is the R-equivalent of an analysis from "Common Statistical
> Methods for Clinical Research with SAS Examples."
> 
> R produces the follwoing output:
> 
>> print(surv.by.vac)
> Call: survfit(formula = Surv(WKS, CENS == 0) ~ VAC, data = hsv)
> 
> records n.max n.start events median 0.95LCL 0.95UCL
> VAC=GD2  2525  25 14 35  15  NA
> VAC=PBO  2323  23 17 15  12  35
> 
> SAS has the same 95% CI for VAC=GD2 but has a 95% CI of [10, 27] for
> VAC=PBO. This is just like in the analysis I'm doing currently.
> 
> Thanks,
> 
> Paul
> 
>  
> ###
>  Chapter 21: The Log-Rank Test 
> ###
>  
> #
>  Example 21.1: HSV2 Vaccine with gD2 Vaccine 
> #
>  
> connection <- textConnection("
> GD2  1   8 12  GD2  3 -12 10  GD2  6 -52  7
> GD2  7  28 10  GD2  8  44  6  GD2 10  14  8
> GD2 12   3  8  GD2 14 -52  9  GD2 15  35 11
> GD2 18   6 13  GD2 20  12  7  GD2 23  -7 13
> GD2 24 -52  9  GD2 26 -52 12  GD2 28  36 13
> GD2 31 -52  8  GD2 33   9 10  GD2 34 -11 16
> GD2 36 -52  6  GD2 39  15 14  GD2 40  13 13
> GD2 42  21 13  GD2 44 -24 16  GD2 46 -52 13
> GD2 48  28  9  PBO  2  15  9  PBO  4 -44 10
> PBO  5  -2 12  PBO  9   8  7  PBO 11  12  7
> PBO 13 -52  7  PBO 16  21  7  PBO 17  19 11
> PBO 19   6 16  PBO 21  10 16  PBO 22 -15  6
> PBO 25   4 15  PBO 27  -9  9  PBO 29  27 10
> PBO 30   1 17  PBO 32  12  8  PBO 35  20  8
> PBO 37 -32  8  PBO 38  15  8  PBO 41   5 14
> PBO 43  35 13  PBO 45  28  9  PBO 47   6 15
> ")
> 
> hsv <- data.frame(scan(connection, list(VAC="", PAT=0, WKS=0, X=0)))
> hsv <- transform(hsv,
>   CENS = ifelse(WKS < 1, 1, 0),
>   WKS  = abs(WKS),
>   TRT  = ifelse(VAC=="GD2", 1, 0))
> 
> library("survival")
> surv.by.vac <- survfit(Surv(WKS,CENS==0)~VAC, data=hsv)
> 
> plot(surv.by.vac, 
>  main = "The Log-Rank Test \n Example 21.1: HSV-Episodes with gD2
> Vaccine",
>  ylab = "Survival Distribution Function",
>  xlab = "Survival Time in Weeks",
>  lty = c(1,2))
> 
> legend(0.75,0.19, 
>  legend = c("gD2","PBO"), 
>  lty = c(1,2), title = "Treatment")
> 
> summary(surv.by.vac)
> print(surv.by.vac)
>  
> 
> __
> 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 University
--
View this message in context: 
http://r.789695.n4.nabble.com/Kaplan-Meier-analysis-95-CI-wider-in-R-than-in-SAS-tp4554559p4555447.html
Sent from the R help mailing list archive at Nabble.com.

__
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] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Paul Miller]

Hi Enrico,

Not sure how SAS builds the CI but I can look into it. The SAS documentation 
does have a section on computational formulas at:

http://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_lifetest_a000259.htm

Although I can't provide my dataset, I can provide the data and code below. 
This is the R-equivalent of an analysis from "Common Statistical Methods for 
Clinical Research with SAS Examples."

R produces the follwoing output:

> print(surv.by.vac)
Call: survfit(formula = Surv(WKS, CENS == 0) ~ VAC, data = hsv)

records n.max n.start events median 0.95LCL 0.95UCL
VAC=GD2  2525  25 14 35  15  NA
VAC=PBO  2323  23 17 15  12  35

SAS has the same 95% CI for VAC=GD2 but has a 95% CI of [10, 27] for VAC=PBO. 
This is just like in the analysis I'm doing currently.

Thanks,

Paul

 
###
 Chapter 21: The Log-Rank Test 
###
 
#
 Example 21.1: HSV2 Vaccine with gD2 Vaccine 
#
 
connection <- textConnection("
GD2  1   8 12  GD2  3 -12 10  GD2  6 -52  7
GD2  7  28 10  GD2  8  44  6  GD2 10  14  8
GD2 12   3  8  GD2 14 -52  9  GD2 15  35 11
GD2 18   6 13  GD2 20  12  7  GD2 23  -7 13
GD2 24 -52  9  GD2 26 -52 12  GD2 28  36 13
GD2 31 -52  8  GD2 33   9 10  GD2 34 -11 16
GD2 36 -52  6  GD2 39  15 14  GD2 40  13 13
GD2 42  21 13  GD2 44 -24 16  GD2 46 -52 13
GD2 48  28  9  PBO  2  15  9  PBO  4 -44 10
PBO  5  -2 12  PBO  9   8  7  PBO 11  12  7
PBO 13 -52  7  PBO 16  21  7  PBO 17  19 11
PBO 19   6 16  PBO 21  10 16  PBO 22 -15  6
PBO 25   4 15  PBO 27  -9  9  PBO 29  27 10
PBO 30   1 17  PBO 32  12  8  PBO 35  20  8
PBO 37 -32  8  PBO 38  15  8  PBO 41   5 14
PBO 43  35 13  PBO 45  28  9  PBO 47   6 15
")

hsv <- data.frame(scan(connection, list(VAC="", PAT=0, WKS=0, X=0)))
hsv <- transform(hsv,
  CENS = ifelse(WKS < 1, 1, 0),
  WKS  = abs(WKS),
  TRT  = ifelse(VAC=="GD2", 1, 0))

library("survival")
surv.by.vac <- survfit(Surv(WKS,CENS==0)~VAC, data=hsv)

plot(surv.by.vac, 
 main = "The Log-Rank Test \n Example 21.1: HSV-Episodes with gD2 Vaccine",
 ylab = "Survival Distribution Function",
 xlab = "Survival Time in Weeks",
 lty = c(1,2))

legend(0.75,0.19, 
 legend = c("gD2","PBO"), 
 lty = c(1,2), title = "Treatment")

summary(surv.by.vac)
print(surv.by.vac)
 

__
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.

[R] Kaplan Meier analysis: 95% CI wider in R than in SAS

[Paul Miller]

Hello All,
 
Am replicating in R an analysis I did earlier using SAS. See this as a test of 
whether I'm ready to start using R in my day-to-day work.
 
Just finished replicating a Kaplan Meier analysis. Everything seems to work out 
fine except for one thing. The 95% CI around my estimate for the median is 
substantially larger in R than in SAS. For example, in SAS I have a median of 
3.29 with a 95% CI of [1.15, 5.29]. In R, I get a median of 3.29 with a 95% CI 
of [1.35, 13.35].
 
Can anyone tell me why I get this difference?
 
My R code looks like:
 
survfrm <- Surv(progression_months_landmark_14,progression==1) ~ 
pr_rg_landmark_14 
survobj <- survfit(survfrm, data=Survival)
survlrk <- survdiff(survfrm, data=Survival)
summary(survobj)
print(survobj)
print(survlrk)
 
My SAS code looks like:
 
proc lifetest data=survival;
strata pr_rg_landmark_14;
time progression_months_landmark_14 * progression(0);
run;

Thought maybe the difference could have something to do with the strata 
statement in the SAS code not being translated properly into R. Tried changing 
my R code to make pr_rg_landmark_14 a strata but this didn't seem to change 
anything. Except that I no longer got a log rank test. 

Thanks,

Paul
 
 
 
 

 
 
 

__
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.