[R] The curious special case of "~ (a + b)/c"
njs at pobox.com
Sat Jun 5 22:30:54 CEST 2010
This isn't at all an urgent practical question, but recently while
exploring the details of how R formulas are interpreted, I learned of
this funny special case for how / interacts with +. In all of the
following cases, the multiplication-like operator simply distributes
(a + b):c = a:c + a:c
a:(b + c) = a:b + a:c
(a + b)*c = a*c + b*c
a*(b + c) = a*b + a*c
a/(b + c) = a/b + a/c
(a + b)/c = a + b + a:b:c, not a/c + b/c = a + a:c + b + b:c
Chambers and Hastie mention this, but give no explanation (page 29/30,
"Slightly more subtle is...").
So for my own edification, does anyone know/care to speculate about
why (a + b)/c works this way?
More information about the R-help