ggcoxdiagnostics warning message

[TeoK]

Hi
I try to use function ggcoxdiagnostics() plots with a parametr ox.scale="time" as provided in survminer cheatsheet

library("survival")
library("survminer")
fit <- coxph(Surv(time, status) ~sex + age, data = lung)
ggcoxdiagnostics(fit, type = "schoenfeld", ox.scale = "time")

And here screenshot from sheet that I would be expected

And here what I really get, with message in console:

Warning messages:
1: In ggcoxdiagnostics(fit_l, type = "schoenfeld", ox.scale = "time") :
NAs introduced by coercion
2: Removed 330 rows containing non-finite values
(stat_smooth).
3: Removed 330 rows containing missing values
(geom_point).

So what wrong? (expect using parametr ecog.ps)

[technocrat]

The syntax is right

library("survival")
library("survminer")
#> Loading required package: ggplot2
#> Loading required package: ggpubr
#> Loading required package: magrittr
lung %>% head(30) -> short_lung
fit <- coxph(Surv(time, status) ~sex + age, data = short_lung)
ggcoxdiagnostics(fit, type = "schoenfeld", ox.scale = "time")

^{Created on 2020-02-06 by the reprex package (v0.3.0)}

So, it must be the rest of the data. Removing age works

library("survival")
library("survminer")
#> Loading required package: ggplot2
#> Loading required package: ggpubr
#> Loading required package: magrittr
lung %>% dplyr::select(time,status,sex,age) -> lung_trim
fit <- coxph(Surv(time, status) ~ sex, data = lung_trim)
ggcoxdiagnostics(fit, type = "schoenfeld", ox.scale = "time")

^{Created on 2020-02-06 by the reprex package (v0.3.0)}

So, consider subsetting out age with NA

[TeoK]

Still it's doesn solve problem, in lung database no NA in Age

[andresrcs]

Hi!

To help us help you, could you please prepare a reproducible example (reprex) illustrating your issue? Please have a look at this guide, to see how to create one:

FAQ: How to do a minimal reproducible example ( reprex ) for beginners

A minimal reproducible example consists of the following items: A minimal dataset, necessary to reproduce the issue The minimal runnable code necessary to reproduce the issue, which can be run on the given dataset, and including the necessary information on the used packages. Let's quickly go over each one of these with examples: Minimal Dataset (Sample Data) You need to provide a data frame that is small enough to be (reasonably) pasted on a post, but big enough to reproduce your issue. Let's say, as an example, that you are working with the iris data frame head(iris) #> Sepal.Length Sepal.Width Petal.Length Petal.Width Species #> 1 5.1 3.5 1.4 0.2…

[nirgrahamuk]

sapply(lung, function(x) sum(is.na(x)))

[TeoK]

sapply(lung, function(x) sum([is.na](http://is.na/)(x)))

Output:

inst      time    status       age       sex 
        1         0         0         0         0 
  ph.ecog  ph.karno pat.karno  meal.cal   wt.loss 
        1         1         3        47        14

so is no NA in "Age" in lung database (provided by survival package)
I just follow by survminer cheatsheet available here
So I still confusing why it's doesn't work

[technocrat]

You're right, as I could have seen by summary(lung).

I noticed that while the cheatsheet did give the same code but didn't show the output.

I looked at the examples in help(ggcoxdiagnostics), one of which was

(coxph.fit2, type = "schoenfeld", ox.scale = "observation.id")

instead of ox.scale = "time")

So, I speculated that changing ox.scale to observation.id would plot, and it does. Now, I speculate that you can time order (but not necessarily scale) simply by sorting on time and get there, since observation.id will be sequential by time.