Hi,
I'm surprised by the behaviour of missmap() when the argument is a dataframe with no
missing data. It seems to plot everything as missing.
I did not see this feature documented in the help. I also didn't find any thread
about that in this mailing list.
I realise this package is not for datasets without missing data, but this may still be
misleading.
Example :
library(package = "Amelia")
tabComplete <- data.frame(a = 1:10, b = 11:20, c = 21:30)
missmap(tabComplete) # everything plotted missing
is.na(tabComplete$a) <- 1
missmap(tabComplete) # works when at least one observation is missing
Best regards,
Antoine Filipovic Pierucci
Institut national d'études démographiques INED
133 Bd Davout - 75 980 Paris 20 - France