Convert to formula class

[zx8754]

Related SO post - https://stackoverflow.com/q/73136299/680068

Where lm works fine and clusWilcox.test fails. lm manual says:

formula - an object of class "formula" (or one that can be coerced to that class)

In clusWilcox.test check if it is not class of formula, then coerce to formula to avoid the error.

[wenjie2wang]

Thanks. It looks like a good suggestion that can be easily applied.

[wenjie2wang]

In brief, this "issue" should be resolved by bd61027.

To be more specific, the clusWilcox.test() is an S3 generic function, with an S3 method for formula objects. The default method will be called if the input x is not a formula object. The substitute(mpg ~ i, list(i = as.name(x)) produces a language object, and thus, the default method is called, which (of course) does not know how to handle the input properly. I do not think it is an issue of the function. But why does lm() work? Notice that lm() is not a generic function. The conversion from language to formula is actually done in model.frame.default().

remotes::install_github("wenjie2wang/clusrank")
library(clusrank)
packageVersion("clusrank")
#> [1] '1.0.3'

## example data
df <- mtcars[, c("vs", "am", "mpg")]
names(df)
#> [1] "vs"  "am"  "mpg"
# [1] "vs"  "am"  "mpg"

varlist <- names(df)[1:2]
# [1] "vs" "am"

lapply(varlist, function(x) {
  clusWilcox.test(substitute(mpg ~ i, list(i = as.name(x))), data = df)
})
#> [[1]]
#> 
#>  Clustered Wilcoxon rank sum test using Rosner-Glynn-Lee method
#> 
#> data:  mpg; group: vs; (from df)
#> number of observations: 32;  number of clusters: 32
#> Z = 3.9342, p-value = 8.349e-05
#> alternative hypothesis: true difference in locations is not equal to 0
#> 
#> 
#> [[2]]
#> 
#>  Clustered Wilcoxon rank sum test using Rosner-Glynn-Lee method
#> 
#> data:  mpg; group: am; (from df)
#> number of observations: 32;  number of clusters: 32
#> Z = 3.1291, p-value = 0.001753
#> alternative hypothesis: true difference in locations is not equal to 0

^{Created on 2022-07-28 by the reprex package (v2.0.1)}