set stringsAsFactors = FALSE in the data frame returned by the function GKtau to be compatible with upcoming change in data frame default for this argument
applied as.character() to GKtau arguments x and y so that missing levels of a factor no longer cause GKtau to return NaN
applied as.data.frame() to GKtauDataframe so that it works correctly with data tables