[R] Problem with paxckage survMisc
Hi R users. I would like your help on the following strange, to me,
behavior of the package survMisc.
I have a simple routine, the following:
proc-function(){
rm(list=ls())
library(survMisc)
d-read.table(C:\\Program
Files\\R\\Data\\Survival\\HosmLem.txt,fill=TRUE,header=TRUE)
d4-as.factor(d[,4])
s-survfit(Surv(d[,2], d[,5])~d4)
ctest-comp(s)$tests
print(ctest)
}
When I run it as a routine I am getting the following message:
Error in Surv(d[, 2], d[, 5]) : object 'd' not found
When I run it command-by-command the program works OK.
Any suggestion what it is going wrong?
The first six lines of the data used are
id Time Age Drug Censorentdateenddate
1 15 460 1 05/15/1990 10/14/1990
2 26 351 0 09/19/1989 03/20/1990
3 38 301 1 04/21/1991 12/20/1991
4 43 301 1 01/03/1991 04/04/1991
5 5 22 360 1 09/18/1989 07/19/1991
6 61 321 0 03/18/1991 04/17/1991
The full data set can be obtained using the command
d-read.table(http://www.ats.ucla.edu/stat/r/examples/asa/hmohiv.csv;,
sep=,, header = TRUE)
Many thanks
Endy
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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] survMisc
Hi R users. I have some problems with the package “survMisc”. When I am
loading it I am getting the following
library(survMisc)
Loading required package: survival
Loading required package: splines
Loading required package: km.ci
Loading required package: ggplot2
Loading required package: data.table
data.table 1.9.4 For help type: ?data.table
*** NB: by=.EACHI is now explicit. See README to restore previous behaviour.
Loading required package: gridExtra
Loading required package: grid
Loading required package: rpart
Attaching package: ‘survMisc’
The following objects are masked from ‘package:stats’:
AIC, BIC, median, quantile
In the above output I noticed the line with the three stars (*). In
order to restore the data.table in its previous behavior I tried to locate
the README file but I couldn’t.
I ignored that NB in the previous output and I continue to run the
example given in the above mentioned package for the routine comp(). The
commands and the output are given below.
### 2 curves
data(kidney,package=KMsurv)
s1 - survfit(Surv(time=time, event=delta) ~ type, data=kidney )
comp(s1)
$tne
t n e n_type=1 e_type=1 n_type=2 e_type=2
1: 1.586 2 431
431
2: 3.580 2 401
401
3: 4.572 4 362
362
4: 5.566 2 331
331
5: 8.560 4 302 30
2
6: 9.554 2 271
271
7: 10.5 50 2 251
251
8: 11.544 2 22 1
221
9: 15.528 4 14 2
142
10: 16.5 26 2 13 1
131
11: 18.5 22 2 11 1
111
12: 23.5 8 24 1
41
13: 26.5 6 23 1
31
$tests
$tests$lrTests
ChiSq df p
Log-rank0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen)0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$tests$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware 0 1
Peto-Peto 0 1
Mod~ Peto-Peto (Andersen) 0 1
Renyi Flem~-Harr~ with p=1, q=1 0 1
Notice the zeros (0) that corresponds to the test statistics. (To my
opinion those zeros are strongly related to the NB above).
Next I noticed the following strange, to my opinion, thing. More
precisely I have written the following
routine
proc-function(){
rm(list=ls())
library(survMisc)
d-read.table(C:\\Program
Files\\R\\Data\\Survival\\HosmLem.txt,fill=TRUE,header=TRUE)
d4-as.factor(d[,4])
s-survfit(Surv(d[,2], d[,5])~d4)
ctest-comp(s)$tests
print(ctest)
}
The data used are those of Hosmer and Lemeshow book on Applied Survival
Analysis. The first rows of this data set follow.
id Time Age Drug Censorentdateenddate
15 460 1 05/15/1990 10/14/1990
26 351 0 09/19/1989 03/20/1990
38 301 1 04/21/1991 12/20/1991
43 301 1 01/03/1991 04/04/1991
5 22 360 1 09/18/1989 07/19/1991
61 321 003/18/1991 04/17/1991
When I run the function proc() I am getting the answer
proc()
Error in Surv(d[, 2], d[, 5]) : object 'd' not found
In contrast when I run the same routine command-by-command I am getting the
following output
$lrTests
ChiSq df p
Log-rank 0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen) 0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware0 1
Peto-Peto0 1
Mod~ Peto-Peto (Andersen) 0 1
[R] survMisc package
Hi R users. I have some problems with the package “survMisc”. When I am
loading it I am getting the following
library(survMisc)
Loading required package: survival
Loading required package: splines
Loading required package: km.ci
Loading required package: ggplot2
Loading required package: data.table
data.table 1.9.4 For help type: ?data.table
*** NB: by=.EACHI is now explicit. See README to restore previous behaviour.
Loading required package: gridExtra
Loading required package: grid
Loading required package: rpart
Attaching package: ‘survMisc’
The following objects are masked from ‘package:stats’:
AIC, BIC, median, quantile
In the above output I noticed the line with the three stars (*). In
order to restore the data.table in its previous behavior I tried to locate
the README file but I couldn’t.
I ignored that NB in the previous output and I continue to run the
example given in the above mentioned package for the routine comp(). The
commands and the output are given below.
### 2 curves
data(kidney,package=KMsurv)
s1 - survfit(Surv(time=time, event=delta) ~ type, data=kidney )
comp(s1)
$tne
t n e n_type=1 e_type=1 n_type=2 e_type=2
1: 1.586 2 431
431
2: 3.580 2 401
401
3: 4.572 4 362
362
4: 5.566 2 331
331
5: 8.560 4 302 30
2
6: 9.554 2 271
271
7: 10.5 50 2 251
251
8: 11.544 2 22 1
221
9: 15.528 4 14 2
142
10: 16.5 26 2 13 1
131
11: 18.5 22 2 11 1
111
12: 23.5 8 24 1
41
13: 26.5 6 23 1
31
$tests
$tests$lrTests
ChiSq df p
Log-rank0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen)0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$tests$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware 0 1
Peto-Peto 0 1
Mod~ Peto-Peto (Andersen) 0 1
Renyi Flem~-Harr~ with p=1, q=1 0 1
Notice the zeros (0) that corresponds to the test statistics. (To my
opinion those zeros are strongly related to the NB above).
Next I noticed the following strange, to my opinion, thing. More
precisely I have written the following
routine
proc-function(){
rm(list=ls())
library(survMisc)
d-read.table(C:\\Program
Files\\R\\Data\\Survival\\HosmLem.txt,fill=TRUE,header=TRUE)
d4-as.factor(d[,4])
s-survfit(Surv(d[,2], d[,5])~d4)
ctest-comp(s)$tests
print(ctest)
}
The data used are those of Hosmer and Lemeshow book on Applied Survival
Analysis. The first rows of this data set follow.
id Time Age Drug Censorentdateenddate
15 460 1 05/15/1990 10/14/1990
26 351 0 09/19/1989 03/20/1990
38 301 1 04/21/1991 12/20/1991
43 301 1 01/03/1991 04/04/1991
5 22 360 1 09/18/1989 07/19/1991
61 321 003/18/1991 04/17/1991
When I run the function proc() I am getting the answer
proc()
Error in Surv(d[, 2], d[, 5]) : object 'd' not found
In contrast when I run the same routine command-by-command I am getting the
following output
$lrTests
ChiSq df p
Log-rank 0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen) 0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware0 1
Peto-Peto0 1
Mod~ Peto-Peto (Andersen) 0 1
[R] survMisc
Hi R users. I have some problems with the package “survMisc”. When I am
loading it I am getting the following
library(survMisc)
Loading required package: survival
Loading required package: splines
Loading required package: km.ci
Loading required package: ggplot2
Loading required package: data.table
data.table 1.9.4 For help type: ?data.table
*** NB: by=.EACHI is now explicit. See README to restore previous behaviour.
Loading required package: gridExtra
Loading required package: grid
Loading required package: rpart
Attaching package: ‘survMisc’
The following objects are masked from ‘package:stats’:
AIC, BIC, median, quantile
In the above output I noticed the line with the three stars (*). In
order to restore the data.table in its previous behavior I tried to locate
the README file but I couldn’t.
I ignored that NB in the previous output and I continue to run the
example given in the above mentioned package for the routine comp(). The
commands and the output are given below.
### 2 curves
data(kidney,package=KMsurv)
s1 - survfit(Surv(time=time, event=delta) ~ type, data=kidney )
comp(s1)
$tne
t n e n_type=1 e_type=1 n_type=2 e_type=2
1: 1.586 2 431
431
2: 3.580 2 401
401
3: 4.572 4 362
362
4: 5.566 2 331
331
5: 8.560 4 302 30
2
6: 9.554 2 271
271
7: 10.5 50 2 251
251
8: 11.544 2 22 1
221
9: 15.528 4 14 2
142
10: 16.5 26 2 13 1
131
11: 18.5 22 2 11 1
111
12: 23.5 8 24 1
41
13: 26.5 6 23 1
31
$tests
$tests$lrTests
ChiSq df p
Log-rank0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen)0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$tests$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware 0 1
Peto-Peto 0 1
Mod~ Peto-Peto (Andersen) 0 1
Renyi Flem~-Harr~ with p=1, q=1 0 1
Notice the zeros (0) that corresponds to the test statistics. (To my
opinion those zeros are strongly related to the NB above).
Next I noticed the following strange, to my opinion, thing. More
precisely I have written the following
routine
proc-function(){
rm(list=ls())
library(survMisc)
d-read.table(C:\\Program
Files\\R\\Data\\Survival\\HosmLem.txt,fill=TRUE,header=TRUE)
d4-as.factor(d[,4])
s-survfit(Surv(d[,2], d[,5])~d4)
ctest-comp(s)$tests
print(ctest)
}
The data used are those of Hosmer and Lemeshow book on Applied Survival
Analysis. The first rows of this data set follow.
id Time Age Drug Censorentdateenddate
15 460 1 05/15/1990 10/14/1990
26 351 0 09/19/1989 03/20/1990
38 301 1 04/21/1991 12/20/1991
43 301 1 01/03/1991 04/04/1991
5 22 360 1 09/18/1989 07/19/1991
61 321 003/18/1991 04/17/1991
When I run the function proc() I am getting the answer
proc()
Error in Surv(d[, 2], d[, 5]) : object 'd' not found
In contrast when I run the same routine command-by-command I am getting the
following output
$lrTests
ChiSq df p
Log-rank 0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen) 0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware0 1
Peto-Peto0 1
Mod~ Peto-Peto (Andersen) 0 1
[R] Package survMisc
Hi R users. I have some problems with the package “survMisc”. When I am
loading it I am getting the following
library(survMisc)
Loading required package: survival
Loading required package: splines
Loading required package: km.ci
Loading required package: ggplot2
Loading required package: data.table
data.table 1.9.4 For help type: ?data.table
*** NB: by=.EACHI is now explicit. See README to restore previous behaviour.
Loading required package: gridExtra
Loading required package: grid
Loading required package: rpart
Attaching package: ‘survMisc’
The following objects are masked from ‘package:stats’:
AIC, BIC, median, quantile
In the above output I noticed the line with the three stars (*). In
order to restore the data.table in its previous behavior I tried to locate
the README file but I couldn’t.
I ignored that NB in the previous output and I continue to run the
example given in the above mentioned package for the routine comp(). The
commands and the output are given below.
### 2 curves
data(kidney,package=KMsurv)
s1 - survfit(Surv(time=time, event=delta) ~ type, data=kidney )
comp(s1)
$tne
t n e n_type=1 e_type=1 n_type=2 e_type=2
1: 1.586 2 431
431
2: 3.580 2 401
401
3: 4.572 4 362
362
4: 5.566 2 331
331
5: 8.560 4 302 30
2
6: 9.554 2 271
271
7: 10.5 50 2 251
251
8: 11.544 2 22 1
221
9: 15.528 4 14 2
142
10: 16.5 26 2 13 1
131
11: 18.5 22 2 11 1
111
12: 23.5 8 24 1
41
13: 26.5 6 23 1
31
$tests
$tests$lrTests
ChiSq df p
Log-rank0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen)0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$tests$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware 0 1
Peto-Peto 0 1
Mod~ Peto-Peto (Andersen) 0 1
Renyi Flem~-Harr~ with p=1, q=1 0 1
Notice the zeros (0) that corresponds to the test statistics. (To my
opinion those zeros are strongly related to the NB above).
Next I noticed the following strange, to my opinion, thing. More
precisely I have written the following
routine
proc-function(){
rm(list=ls())
library(survMisc)
d-read.table(C:\\Program
Files\\R\\Data\\Survival\\HosmLem.txt,fill=TRUE,header=TRUE)
d4-as.factor(d[,4])
s-survfit(Surv(d[,2], d[,5])~d4)
ctest-comp(s)$tests
print(ctest)
}
The data used are those of Hosmer and Lemeshow book on Applied Survival
Analysis. The first rows of this data set follow.
id Time Age Drug Censorentdateenddate
15 460 1 05/15/1990 10/14/1990
26 351 0 09/19/1989 03/20/1990
38 301 1 04/21/1991 12/20/1991
43 301 1 01/03/1991 04/04/1991
5 22 360 1 09/18/1989 07/19/1991
61 321 003/18/1991 04/17/1991
When I run the function proc() I am getting the answer
proc()
Error in Surv(d[, 2], d[, 5]) : object 'd' not found
In contrast when I run the same routine command-by-command I am getting the
following output
$lrTests
ChiSq df p
Log-rank 0 1 1
Gehan-Breslow (mod~ Wilcoxon) 0 1 1
Tarone-Ware 0 1 1
Peto-Peto 0 1 1
Mod~ Peto-Peto (Andersen) 0 1 1
Flem~-Harr~ with p=1, q=1 0 1 1
$supTests
Q p
Log-rank 0 1
Gehan-Breslow (mod~ Wilcoxon) 0 1
Tarone-Ware0 1
Peto-Peto0 1
Mod~ Peto-Peto (Andersen) 0 1
[R] Programming routine comp()
Dear R users. I am trying to program the comp() routine in package survMisc. I am reading the data below with d=read.table( C:\\. . .,fill=TRUE,header=TRUE) Then I load the packages 'survival' and 'survMisc', library(survival), library(survMisc) and I run the commands s=survfit(Surv(d[,2], d[,3])~d[,1], data=d) comp(s) and I am getting the error Error in get(t1, loc1) : object 'd[, 2]' not found If instead I use the commands s=survfit(Surv(T, Status)~Group, data=d) comp(s) routine comp() runs perfectly. However when I am programing I can't see a way to know in advance the variable names in order to use them. Can anybody give me a suggestion? Thanks in advance Endy NB. The data must be stacked in three (3) columns before red. They are repeated in nine (9) columns for space saving. Group T Status Group T Status Group T Status 1 2081 0 1 55 1 2 414 1 1 1602 0 1 1 1 2 2204 1 1 1496 0 1 107 1 2 1063 1 1 1462 0 1 110 1 2 481 1 1 1433 0 1 332 1 2 105 1 1 1377 0 2 2569 0 2 641 1 1 1330 0 2 2506 0 2 390 1 1 996 0 2 2409 0 2 288 1 1 226 0 2 2218 0 2 421 1 1 1199 0 2 1857 0 2 79 1 1 0 2 1829 0 2 748 1 1 530 0 2 1562 0 2 486 1 1 1182 0 2 1470 0 2 48 1 1 1167 0 2 1363 0 2 272 1 1 418 1 2 1030 0 2 1074 1 1 383 1 2 860 0 2 381 1 1 276 1 2 1258 0 2 10 1 1 104 1 2 2246 0 2 53 1 1 609 1 2 1870 0 2 80 1 1 172 1 2 1799 0 2 35 1 1 487 1 2 1709 0 2 248 1 1 662 1 2 1674 0 2 704 1 1 194 1 2 1568 0 2 211 1 1 230 1 2 1527 0 2 219 1 1 526 1 2 1324 0 2 606 1 1 122 1 2 957 0 1 129 1 2 932 0 1 74 1 2 847 0 1 122 1 2 848 0 1 86 1 2 1850 0 1 466 1 2 1843 0 1 192 1 2 1535 0 1 109 1 2 1447 0 1 55 1 2 1384 0 [[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.
[R] Programming routine comp()
Dear R users. I am trying to program the comp() routine in package survMisc. I am reading the data below with d=read.table( C:\\. . .,fill=TRUE,header=TRUE) Then I load the packages 'survival' and 'survMisc', library(survival), library(survMisc) and I run the commands s=survfit(Surv(d[,2], d[,3])~d[,1], data=d) comp(s) and I am getting the error Error in get(t1, loc1) : object 'd[, 2]' not found If instead I use the commands s=survfit(Surv(T, Status)~Group, data=d) comp(s) routine comp() runs perfectly. However when I am programing I can't see a way to know in advance the variable names in order to use them. Can anybody give me a suggestion? Thanks in advance Endy NB. The data must be stacked in three (3) columns before red. They are repeated in nine (9) columns for space saving. Group T Status Group T Status Group T Status 1 2081 0 1 55 1 2 414 1 1 1602 0 1 1 1 2 2204 1 1 1496 0 1 107 1 2 1063 1 1 1462 0 1 110 1 2 481 1 1 1433 0 1 332 1 2 105 1 1 1377 0 2 2569 0 2 641 1 1 1330 0 2 2506 0 2 390 1 1 996 0 2 2409 0 2 288 1 1 226 0 2 2218 0 2 421 1 1 1199 0 2 1857 0 2 79 1 1 0 2 1829 0 2 748 1 1 530 0 2 1562 0 2 486 1 1 1182 0 2 1470 0 2 48 1 1 1167 0 2 1363 0 2 272 1 1 418 1 2 1030 0 2 1074 1 1 383 1 2 860 0 2 381 1 1 276 1 2 1258 0 2 10 1 1 104 1 2 2246 0 2 53 1 1 609 1 2 1870 0 2 80 1 1 172 1 2 1799 0 2 35 1 1 487 1 2 1709 0 2 248 1 1 662 1 2 1674 0 2 704 1 1 194 1 2 1568 0 2 211 1 1 230 1 2 1527 0 2 219 1 1 526 1 2 1324 0 2 606 1 1 122 1 2 957 0 1 129 1 2 932 0 1 74 1 2 847 0 1 122 1 2 848 0 1 86 1 2 1850 0 1 466 1 2 1843 0 1 192 1 2 1535 0 1 109 1 2 1447 0 1 55 1 2 1384 0 [[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.
[R] Programing routine comp()
Dear R users. I am trying to program the comp() routine in package survMisc. I am reading the data below with d=read.table( C:\\. . .,fill=TRUE,header=TRUE) Then I load the packages 'survival' and 'survMisc', library(survival), library(survMisc) and I run the commands s=survfit(Surv(d[,2], d[,3])~d[,1], data=d) comp(s) and I am getting the error Error in get(t1, loc1) : object 'd[, 2]' not found If instead I use the commands s=survfit(Surv(T, Status)~Group, data=d) comp(s) routine comp() runs perfectly. However when I am programing I can't see a way to know in advance the variable names in order to use them. Can anybody give me a suggestion? Thanks in advance Endy NB. The data must be stacked in three (3) columns before red. They are repeated in nine (9) columns for space saving. Group T Status Group T Status Group T Status 1 2081 0 1 55 1 2 414 1 1 1602 0 1 1 1 2 2204 1 1 1496 0 1 107 1 2 1063 1 1 1462 0 1 110 1 2 481 1 1 1433 0 1 332 1 2 105 1 1 1377 0 2 2569 0 2 641 1 1 1330 0 2 2506 0 2 390 1 1 996 0 2 2409 0 2 288 1 1 226 0 2 2218 0 2 421 1 1 1199 0 2 1857 0 2 79 1 1 0 2 1829 0 2 748 1 1 530 0 2 1562 0 2 486 1 1 1182 0 2 1470 0 2 48 1 1 1167 0 2 1363 0 2 272 1 1 418 1 2 1030 0 2 1074 1 1 383 1 2 860 0 2 381 1 1 276 1 2 1258 0 2 10 1 1 104 1 2 2246 0 2 53 1 1 609 1 2 1870 0 2 80 1 1 172 1 2 1799 0 2 35 1 1 487 1 2 1709 0 2 248 1 1 662 1 2 1674 0 2 704 1 1 194 1 2 1568 0 2 211 1 1 230 1 2 1527 0 2 219 1 1 526 1 2 1324 0 2 606 1 1 122 1 2 957 0 1 129 1 2 932 0 1 74 1 2 847 0 1 122 1 2 848 0 1 86 1 2 1850 0 1 466 1 2 1843 0 1 192 1 2 1535 0 1 109 1 2 1447 0 1 55 1 2 1384 0 [[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.
[R] How I can program comp() routine
Group T Status Dear R users. I am trying to program the comp() routine in package survMisc. 1 2081 0 I am reading the data on the left with d=read.table( C:\\e.t.c,fill=TRUE,header=TRUE) 1 1602 0 Then I load the packages 'survival' and 'survMisc', library(survival), library(survMisc) 1 1496 0 and I run the commands 1 1462 0 s=survfit(Surv(d[,2], d[,3])~d[,1], data=d) 1 1433 0 comp(s) 1 1377 0 and I am getting the error 1 1330 0 Error in get(t1, loc1) : object 'd[, 2]' not found 1 996 0 If instead I use the commands 1 226 0s=survfit(Surv(T, Status)~Group, data=d) 1 1199 0 comp(s) 1 0routine comp() runs perfectly. However when I am programing I can't see a way to know 1 530 0 in advance the variable names in order to use them. 1 1182 0Can anybody give me a suggestion? 1 1167 0 Thanks in advance 1 418 1 Endy 1 383 1 1 276 1 1 104 1 1 609 1 1 172 1 1 487 1 1 662 1 1 194 1 1 230 1 1 526 1 1 122 1 1 129 1 1 74 1 1 122 1 1 86 1 1 466 1 1 192 1 1 109 1 1 55 1 11 1 1 107 1 1 110 1 1 332 1 2 2569 0 2 2506 0 2 2409 0 2 2218 0 2 1857 0 2 1829 0 2 1562 0 2 1470 0 2 1363 0 2 1030 0 2 860 0 2 1258 0 2 2246 0 2 1870 0 2 1799 0 2 1709 0 2 1674 0 2 1568 0 2 1527 0 2 1324 0 2 957 0 2 932 0 2 847 0 2 848 0 2 1850 0 2 1843 0 2 1535 0 2 1447 0 2 1384 0 2 414 1 2 2204 1 2 1063 1 2 481 1 2 105 1 2 641 1 2 390 1 2 288 1 2 421 1 2 79 1 2 748 1 2 486 1 2 48 1 2 272 1 2 1074 1 2 381 1 2 10 1 2 53 1 2 80 1 2 35 1 2 248 1 2 704 1 2 211 1 2 219 1 [[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.
[R] Survival analysis
Dear R users, how I can implement Breslow and Tarone-Ware tests for survival analysis? Also, is there any way I can estimate and plot the hazard function,along the lines of the survival function in Survival Package? (The package muhaz seems that does not do what it promises, except if I could not understand how it works.) [[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.
[R] log-log link function
Hi to every body. I would like assistance on how to implement the log-log link function for binary response. Is there any package that implements it? Many thanks Endy [[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.
[R] Hosmer Lemeshow test
Hi to everybody. I use the following routine (i found it in the internet)
to compute the Hosmer-Lemeshow test in the framework of logistic regression.
hosmerlemeshow = function(obj, g=10) {
# first, check to see if we fed in the right kind of object
stopifnot(family(obj)$family==binomial family(obj)$link==logit)
y = obj$model[[1]]
# the double bracket (above) gets the index of items within an object
if (is.factor(y))
y = as.numeric(y)==2
yhat = obj$fitted.values
cutyhat=cut(yhat,quantile(yhat,0:g/g),include.lowest=TRUE)
obs = xtabs(cbind(1 - y, y) ~ cutyhat)
expect = xtabs(cbind(1 - yhat, yhat) ~ cutyhat)
if (any(expect 5))
# warning(Some expected counts are less than 5. Use smaller number of
groups)
chisq = sum((obs - expect)^2/expect)
P = 1 - pchisq(chisq, g - 2)
cat(H-L2 statistic,round(chisq,4), and its p value,P,\n)
# by returning an object of class htest, the function will perform
like the
# built-in hypothesis tests
return(structure(list(
method = c(paste(Hosmer and Lemeshow goodness-of-fit test with, g,
bins, sep= )),
data.name = deparse(substitute(obj)),
statistic = c(X2=chisq),
parameter = c(df=g-2),
p.value = P
), class='htest'))
return(list(chisq=chisq,p.value=P))
}
However when i run it with the data set below i get the value NaN. Using
the same data set with SPSS i get a specific value.
FlightNo Temp ThermalDisast
1 66 0
2 70 1
3 69 0
4 68 0
5 67 0
6 72 0
7 73 0
8 70 0
9 57 1
10 63 1
11 70 1
12 78 0
13 67 0
14 53 1
15 67 0
16 75 0
17 70 0
18 81 0
19 76 0
20 79 0
21 75 1
22 76 0
23 58 1
Any suggestions will greatly appreciated.
With regards
Endy
[[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.
[R] Logistic regression
Dear colleagues I have a couple of problems related with binary logistic
regression.
The first problem is how to compute Pearson and Likelihood chi-squeared
tests for grouped data.
For the same form of data set how to compute sensitivity, specificity and
related measures.
When I speak about grouped data I mean data of the following form
Alcohol.Consum Malformed NoMalformed
0 4817066
1 3814464
1-2 5 788
3-5 1 126
=6 1 37
(This data set has been taken from the book Categorical data analysis by
Agresti)
The second question is the following:is it correct to upload a grouped data
set to an ungrouped one?
The upload is achieved with the aid of the following routine
Createdataframe - function(d){
f1 = f2 = numeric(0)
v = d[,1]
for(j in 2:3){
for(i in 1:dim(d)[1]){
f1 = c( f1, rep( levels(v)[v[i]], d[i,j] ) )
}
f2 = c( f2, rep( 3-j, sum(d[,j]) ) )
}
df=data.frame(f1,f2)
return(df)
}
My finally question is why SPSS computes the Hosmer-Lemeshow test while the
routine listed below gives NaN. (One has to put g=8 in order to get a
numeric value)
The data set is (it is also has been taken from the same book mentioned
above)
FlightNo Temp ThermalDisast
1 66 0
2 70 1
3 69 0
4 68 0
5 67 0
6 72 0
7 73 0
8 70 0
9 57 1
10 63 1
11 70 1
12 78 0
13 67 0
14 53 1
15 67 0
16 750
17 700
18 810
19 760
20 790
21 751
22 760
23 58 1
and the routine used is
hosmerlemeshow = function(obj, g=10) {
# first, check to see if we fed in the right kind of object
stopifnot(family(obj)$family==binomial family(obj)$link==logit)
y = obj$model[[1]]
# the double bracket (above) gets the index of items within an object
if (is.factor(y))
y = as.numeric(y)==2
yhat = obj$fitted.values
cutyhat=cut(yhat,quantile(yhat,0:g/g),include.lowest=TRUE)
obs = xtabs(cbind(1 - y, y) ~ cutyhat)
expect = xtabs(cbind(1 - yhat, yhat) ~ cutyhat)
if (any(expect 5))
# warning(Some expected counts are less than 5. Use smaller number
of groups)
chisq = sum((obs - expect)^2/expect)
P = 1 - pchisq(chisq, g - 2)
cat(H-L2 statistic,round(chisq,4), and its p value,P,\n)
# by returning an object of class htest, the function will
perform like the
# built-in hypothesis tests
return(structure(list(
method = c(paste(Hosmer and Lemeshow goodness-of-fit test with, g,
bins, sep= )),
data.name = deparse(substitute(obj)),
statistic = c(X2=chisq),
parameter = c(df=g-2),
p.value = P
), class='htest'))
return(list(chisq=chisq,p.value=P))
}
I found this routine in the internet.
Thank you for your cooperation in advance. Any suggestion and/or solution
will be greatly appreciated.
With regards
Endy
[[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.
[R] Logistic Regression
Please read the attach file. Thank you Endy __ 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] Logistic Regression
Dear colleagues I have a couple of problems related with binary logistic
regression.
The first problem is how to compute Pearson and Likelihood chi-squeared
tests for grouped data.
For the same form of data set how to compute sensitivity, specificity and
related measures.
When I speak about grouped data I mean data of the following form
Alcohol.Consum Malformed NoMalformed
0
4817066
1
3814464
1-2
5788
3-5
1126
=6
1 37
(This data set has been taken from the book Categorical data analysis by
Agresti)
The second question is the following:is it correct to upload a grouped
data set to an ungrouped one?
The upload is achieved with the aid of the following routine
Createdataframe - function(d){
f1 = f2 = numeric(0)
v = d[,1]
for(j in 2:3){
for(i in 1:dim(d)[1]){
f1 = c( f1, rep( levels(v)[v[i]], d[i,j] ) )
}
f2 = c( f2, rep( 3-j, sum(d[,j]) ) )
}
df=data.frame(f1,f2)
return(df)
}
My finally question is why SPSS computes the Hosmer-Lemeshow test while
the routine listed below gives NaN. (One has to put g=8 in order to get a
numeric value)
The data set is (it is also has been taken from the same book mentioned
above)
FlightNo TempThermalDisast
1 660
2 701
3 690
4 680
5 670
6 720
7 730
8 700
9 571
10631
11701
12780
13670
14531
15670
16750
17700
18810
19760
20790
21751
22760
23581
and the routine used is
hosmerlemeshow = function(obj, g=10) {
# first, check to see if we fed in the right kind of object
stopifnot(family(obj)$family==binomial family(obj)$link==logit)
y = obj$model[[1]]
# the double bracket (above) gets the index of items within an object
if (is.factor(y))
y = as.numeric(y)==2
yhat = obj$fitted.values
cutyhat=cut(yhat,quantile(yhat,0:g/g),include.lowest=TRUE)
obs = xtabs(cbind(1 - y, y) ~ cutyhat)
expect = xtabs(cbind(1 - yhat, yhat) ~ cutyhat)
if (any(expect 5))
# warning(Some expected counts are less than 5. Use smaller number of
groups)
chisq = sum((obs - expect)^2/expect)
P = 1 - pchisq(chisq, g - 2)
cat(H-L2 statistic,round(chisq,4), and its p value,P,\n)
# by returning an object of class htest, the function will perform
like the
# built-in hypothesis tests
return(structure(list(
method = c(paste(Hosmer and Lemeshow goodness-of-fit test with, g,
bins, sep= )),
data.name = deparse(substitute(obj)),
statistic = c(X2=chisq),
parameter = c(df=g-2),
p.value = P
), class='htest'))
return(list(chisq=chisq,p.value=P))
}
Thank you for your cooperation in advance. Any suggestion and/or solution
will be greatly appreciated.
With regards
Endy
[[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.
[R] Logistic regression
I have a data set to be analyzed using to binary logistic regression. The data set is iin grouped form. My question is: how I can compute Hosmer-Lemeshow test and measures like sensitivity and specificity? Any suggestion will be greatly appreciated. Thank you Endy [[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.
[R] sensitivity for grouped data
I could not locate any package (using RSeek.org) to compute sensitivity, specificity and related measures, for logistic regression models with grouped data. (For ungrouped data I know one, the SMDTools). Any suggestion will be greatly appreciated. Many thanks Endy [[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.
[R] Help on qpcR package
I am using R on a Windows XP professional platform.
The following code is part of a bigger one
CODE
press=function(y,x){
library(qpcR)
models.press=numeric(0)
cat(\n)
dep=y
print(dep)
indep=log(x)
print(indep)
yfit=dep-PRESS(lm(dep~indep))[[2]]
cat(\n yfit\n)
print(yfit)
yfit.orig=yfit
presid=y-yfit.orig
press=sum(presid^2)
cat(\n)
cat(PRESS =,press,\n)
}
On the command R window I define
- Ignored:
x=c(12,24,13,11,23,10,9,17,11,14,18)
y=c(10,11,9,8,5,12,11,21,12,13,14)
then I load the press source code, I run it using the x and y values
defined before
and I get the following
press(x,y)
Loading required package: drc
Loading required package: alr3
Loading required package: lattice
Loading required package: magic
Loading required package: abind
Loading required package: MASS
Loading required package: nlme
Loading required package: plotrix
'drc' has been loaded.
Please cite R and 'drc' if used for a publication,
for references type 'citation()' and 'citation('drc')'.
Loading required package: gtools
Loading required package: gplots
Loading required package: gdata
Loading required package: caTools
Loading required package: bitops
Loading required package: grid
Attaching package: 'gplots'
The following object(s) are masked from package:plotrix :
plotCI
The following object(s) are masked from package:stats :
lowess
'qpcR' has been loaded.
Please cite R and the following if used for a publication:
Spiess AN, Feig C, Ritz
Highly accurate sigmoidal fitting of real-time PCR data by introducing
a parameter for asymmetry.
BMC Bioinformatics 2008, 29:221
or
Ritz C, Spiess AN. qpcR: an R package for sigmoidal model selection in
quantitative real-time polymerase chain reaction analysis.
Bioinformatics 2008, 24:1549-1551
Newest version always available at www.dr-spiess.de/qpcR.html.
Attaching package: 'qpcR'
The following object(s) are masked from package:gplots :
residplot
[1] 12 24 13 11 23 10 9 17 11 14 18
[1] 2.302585 2.397895 2.197225 2.079442 1.609438 2.484907 2.397895
3.044522 2.484907 2.564949 2.639057
Error in get(noquote(a)) : object 'dep' not found
The error reported, as I understand it, is that the object dep it is
not defined. Obviously, at least to me, it is defined.
If I declare the dep and indep objects on the command R window i. e.
dep=y
indep=x
and rerun the programme then it runs correctly and gives
press(x,y)
[1] 12 24 13 11 23 10 9 17 11 14 18
[1] 2.302585 2.397895 2.197225 2.079442 1.609438 2.484907 2.397895
3.044522 2.484907 2.564949 2.639057
yfit
[1] 13.764062 24.199292 15.804140 15.192045 32.465326 9.547579
9.845243 6.176015 10.521896 12.325493 14.931854
PRESS = 248.7043
Can anybody point out what's the problem?
Thanks in advance
K. Karakostas
__
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.
