In R 2.8.0/Windows:
> lag(1)
[1] 1
attr(,"tsp")
[1] 0 0 1
Though this has the 'tsp' attribute, it is of class "integer",
and hence not
a "time-series object" as defined by ? ts: "time-series
objects... are
vector [sic] or matrices with class of '"ts"' (and additional
attributes)".
? lag further says "Vector or matrix arguments 'x' are coerced to
time
series", but
> identical(lag(as.ts(1)),lag(1))
[1] FALSE
I believe this can be fixed by adding class(x)<-"ts" after line 7
of
stats:::lag.default (x <- hasTsp(x)).
Perhaps all other users of hasTsp expect the same behavior? in which case it
would better be fixed there.
-s
[[alternative HTML version deleted]]