I have data as like the following:
data = data.frame( id = rbinom(1000, 10, .75),
visit = sample(1:3, 1000, replace = TRUE),
room = sample(letters[1:5], 1000, replace = TRUE),
value = rnorm(1000, 50, 10),
timepoint = abs(rnorm(1000))
)
head(data)
id visit room value timepoint
1 8 3 a 62.53394 1.64681140
2 9 1 c 53.67313 1.04093204
3 6 1 c 64.96674 0.40599449
4 8 2 d 41.04145 0.09911475
5 7 2 b 63.86938 1.01732424
6 7 3 c 42.03524 2.04128413
I have defined a S4 class to read this data as longitudinalData. The class goes as follows.
setClass("longitudinalData",
slots = list(id = "integer",
visit = "integer",
room = "character",
value = "numeric",
timepoint = 'numeric'))
To initiate a new object in this class I have defined the following function.
make_LD = function(x){
new("longitudinalData",
id = x$id,
visit = x$visit,
room = x$room,
value = x$value,
timepoint = x$timepoint
)
}
when I try add a new object by make_LD(data) I get the following error.
> make_LD(data)
Error in initialize(value, ...) :
no slot of name "refMethods" for this object of class "classRepresentation"
The easiest way to avoid these sorts of problems is to save the value from setClass, which returns a handy constructor function for you to use.
setClass("longitudinalData",
slots = list(id = "integer",
visit = "integer",
room = "character",
value = "numeric",
timepoint = 'numeric')) -> longitudinalData
make_LD = function(x){
longitudinalData(
id = x$id,
visit = x$visit,
room = x$room,
value = x$value,
timepoint = x$timepoint
)
}
It is normal to give the constructor function the same name as the class, but you don't have to.