Split a data frame into multiple new data frames based on one or
two grouping variables. The surv_group_by() function takes an
existing data frame and converts it into a grouped data frame where
survival analysis are performed "by group".
surv_group_by(data, grouping.vars)
| data | a data frame |
|---|---|
| grouping.vars | a character vector containing the name of grouping variables. Should be of length <= 2 |
Returns an object of class surv_group_by which is a
tibble data frame with the following components:
one column for each grouping variables. Contains the levels.
a coumn named "data", which is a named list of data subsets created by the grouping variables. The list names are created by concatening the levels of grouping variables.
library("survival") library("magrittr") # Grouping by one variables: treatment "rx" #:::::::::::::::::::::::::::::::::::::::::: grouped.d <- colon %>% surv_group_by("rx") grouped.d # print#> # A tibble: 3 x 2 #> # Groups: rx [3] #> rx data #> * <fct> <named list> #> 1 Obs <tibble [630 × 15]> #> 2 Lev <tibble [620 × 15]> #> 3 Lev+5FU <tibble [608 × 15]>grouped.d$data # Access to the data#> $rx.Obs #> # A tibble: 630 x 15 #> id study sex age obstruct perfor adhere nodes status differ extent #> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 3 1 0 71 0 0 1 7 1 2 2 #> 2 3 1 0 71 0 0 1 7 1 2 2 #> 3 5 1 1 69 0 0 0 22 1 2 3 #> 4 5 1 1 69 0 0 0 22 1 2 3 #> 5 8 1 1 54 0 0 0 1 0 2 3 #> 6 8 1 1 54 0 0 0 1 0 2 3 #> 7 13 1 1 64 0 0 0 1 1 2 3 #> 8 13 1 1 64 0 0 0 1 1 2 3 #> 9 15 1 1 46 1 0 0 4 0 2 3 #> 10 15 1 1 46 1 0 0 4 0 2 3 #> # … with 620 more rows, and 4 more variables: surg <dbl>, node4 <dbl>, #> # time <dbl>, etype <dbl> #> #> $rx.Lev #> # A tibble: 620 x 15 #> id study sex age obstruct perfor adhere nodes status differ extent #> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 7 1 1 77 0 0 0 5 1 2 3 #> 2 7 1 1 77 0 0 0 5 1 2 3 #> 3 9 1 1 46 0 0 1 2 0 2 3 #> 4 9 1 1 46 0 0 1 2 0 2 3 #> 5 11 1 0 47 0 0 1 1 0 2 3 #> 6 11 1 0 47 0 0 1 1 0 2 3 #> 7 14 1 1 68 1 0 0 3 1 2 3 #> 8 14 1 1 68 1 0 0 3 1 2 3 #> 9 17 1 1 62 1 0 1 6 1 2 3 #> 10 17 1 1 62 1 0 1 6 1 2 3 #> # … with 610 more rows, and 4 more variables: surg <dbl>, node4 <dbl>, #> # time <dbl>, etype <dbl> #> #> $`rx.Lev+5FU` #> # A tibble: 608 x 15 #> id study sex age obstruct perfor adhere nodes status differ extent #> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 1 1 1 43 0 0 0 5 1 2 3 #> 2 1 1 1 43 0 0 0 5 1 2 3 #> 3 2 1 1 63 0 0 0 1 0 2 3 #> 4 2 1 1 63 0 0 0 1 0 2 3 #> 5 4 1 0 66 1 0 0 6 1 2 3 #> 6 4 1 0 66 1 0 0 6 1 2 3 #> 7 6 1 0 57 0 0 0 9 1 2 3 #> 8 6 1 0 57 0 0 0 9 1 2 3 #> 9 10 1 0 68 0 0 0 1 0 2 3 #> 10 10 1 0 68 0 0 0 1 0 2 3 #> # … with 598 more rows, and 4 more variables: surg <dbl>, node4 <dbl>, #> # time <dbl>, etype <dbl> #># Grouping by two variables #:::::::::::::::::::::::::::::::::::::::::: grouped.d <- colon %>% surv_group_by(grouping.vars = c("rx", "adhere")) grouped.d#> # A tibble: 6 x 3 #> # Groups: rx, adhere [6] #> rx adhere data #> * <fct> <dbl> <named list> #> 1 Obs 0 <tibble [536 × 14]> #> 2 Obs 1 <tibble [94 × 14]> #> 3 Lev 0 <tibble [522 × 14]> #> 4 Lev 1 <tibble [98 × 14]> #> 5 Lev+5FU 0 <tibble [530 × 14]> #> 6 Lev+5FU 1 <tibble [78 × 14]>