Hello,
Is this intended:
> m <- data.frame( date = seq.Date( as.Date("2000-01-01"),
as.Date("2000-01-05"), by = "day" ) )
> sapply( m, typeof )
date
"double"
> as.matrix( m )
date
[1,] "2000-01-01"
[2,] "2000-01-02"
[3,] "2000-01-03"
[4,] "2000-01-04"
[5,] "2000-01-05"
> typeof( as.matrix( m ) )
[1] "character"
In particular :
> m <- data.frame( date = seq.Date( as.Date("2000-01-01"), by = "day",
length.out = 50 ), x = rnorm(50) )
> loess( x ~ date, data= m)
Error: NA/NaN/Inf in foreign function call (arg 2)
In addition: Warning message:
NAs introduced by coercion
> loess( x ~ as.double(date), data= m)
Call:
loess(formula = x ~ as.double(date), data = m)
Number of Observations: 50
Equivalent Number of Parameters: 4.42
Residual Standard Error: 0.9712
Replacing this line in loess :
x <- as.matrix(x)
by this :
x <- tryCatch( do.call( cbind, lapply( x, as.double ) ), error =
function(e){
stop( "predictors must all be numeric" )
} )
does the trick for me.
Romain