R help - Jan 2012 - extracting from data.frames for survival analysis

Hi,

I have a data frame:
> class(B27.vec)
[1] "data.frame"
> head(B27.vec)
  AGE Gend B27 AgeOn DD uveitis psoriasis IBD CD UC InI BASDAI BASFI Smok UV
1  57    1   1    19 38       2         1   1  1  1   1   5.40  8.08   NA  1
2  35    1   1    33  2       2         1   1  1  1   1   1.69  2.28   NA  1
3  49    2   1    40  9       1         1   1  1  1   1   8.30  9.40   NA  0
4  32    1   1    21 11       1         1   1  1  1   1   5.10  9.10   NA  0
5  31    1   1    24  7       1         1   1  1  1   1   6.63  6.52   NA  0
6  27    1   1    23  4       1         2   1  1  1   1   7.19  6.51   NA  0

I am trying to perform survival analysis but continually get errors
when extracting from this data.frame:

attempt 1:
> X <- Surv(B27.vec$AgeOn,B27.vec$UV)
> survdiff(X,rho=0,data=uvf)
Error in x$terms : $ operator is invalid for atomic vectors

attempt 2:
> X <- Surv(B27.vec[,4],B27.vec[,15])
> survdiff(X,rho=0,data=uvf)
Error in x$terms : $ operator is invalid for atomic vector

attempt 3:
> AO <- B27.vec[["AgeOn", exact = TRUE]]
> UV <- B27.vec[["UV",exact=TRUE]]
> X <- Surv(AO,UV)
> survdiff(X,rho=0,data=uvf)
Error in x$terms : $ operator is invalid for atomic vectors

I have read ?data.frame & extract.data.frame but I cannot understand
how I might structure this differently so it extracts the required
columns from this dataframe. For the second 2 attempts I am not using
the $ term. Sorry if this seems basic but cannot understand why
attempt 1 or 2 doesn't work.

thanks
Philip

Hi, Philip,

counter-questions:

1. Which/where is the grouping variable for the test of differences in 
survival?

2. Assume the grouping variable is Gend in B27.vec. Then, why aren't you 
using

survdiff( Surv( AgeOn, UV) ~ Gend, rho = 0, data = B27.vec)

?

Hth  -- Gerrit

On Thu, 26 Jan 2012, Philip Robinson wrote:
> Hi,
>
> I have a data frame:
>
>> class(B27.vec)
> [1] "data.frame"
>
>> head(B27.vec)
>
>  AGE Gend B27 AgeOn DD uveitis psoriasis IBD CD UC InI BASDAI BASFI Smok UV
> 1  57    1   1    19 38       2         1   1  1  1   1   5.40  8.08   NA 
1
> 2  35    1   1    33  2       2         1   1  1  1   1   1.69  2.28   NA 
1
> 3  49    2   1    40  9       1         1   1  1  1   1   8.30  9.40   NA 
0
> 4  32    1   1    21 11       1         1   1  1  1   1   5.10  9.10   NA 
0
> 5  31    1   1    24  7       1         1   1  1  1   1   6.63  6.52   NA 
0
> 6  27    1   1    23  4       1         2   1  1  1   1   7.19  6.51   NA 
0
>
> I am trying to perform survival analysis but continually get errors
> when extracting from this data.frame:
>
> attempt 1:
>> X <- Surv(B27.vec$AgeOn,B27.vec$UV)
>> survdiff(X,rho=0,data=uvf)
> Error in x$terms : $ operator is invalid for atomic vectors
>
> attempt 2:
>> X <- Surv(B27.vec[,4],B27.vec[,15])
>> survdiff(X,rho=0,data=uvf)
> Error in x$terms : $ operator is invalid for atomic vector
>
> attempt 3:
>> AO <- B27.vec[["AgeOn", exact = TRUE]]
>> UV <- B27.vec[["UV",exact=TRUE]]
>> X <- Surv(AO,UV)
>> survdiff(X,rho=0,data=uvf)
> Error in x$terms : $ operator is invalid for atomic vectors
>
> I have read ?data.frame & extract.data.frame but I cannot understand
> how I might structure this differently so it extracts the required
> columns from this dataframe. For the second 2 attempts I am not using
> the $ term. Sorry if this seems basic but cannot understand why
> attempt 1 or 2 doesn't work.
>
> thanks
> Philip
>
> ______________________________________________
> R-help at 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.
>

--- begin included message ---
...
I am trying to perform survival analysis but continually get errors
when extracting from this data.frame:

attempt 1:
> X <- Surv(B27.vec$AgeOn,B27.vec$UV)
> survdiff(X,rho=0,data=uvf)
Error in x$terms : $ operator is invalid for atomic vectors

...
--------- end inclusion ----

As should be apparent from Gerrit Eichner's response, your problem is
that survdiff expects a formula as the first argument and you gave it a
survival object instead.
  However, the error message you got is really not helpful.  I've now
added a check to survreg that will result in a more apropos response
from the routine.

Terry Therneau