| To: | "S-PLUS Newsgroup (E-mail)" <s-news@lists.biostat.wustl.edu> |
|---|---|
| Subject: | Summary: tapply() |
| From: | Winnie Lambert <lambert.winifred@ensco.com> |
| Date: | Thu, 13 Dec 2001 17:00:00 -0500 |
|
All,
Sincere thanks to
all who suggested to make $Hour a factor:
Russel
Zaretzki
Samuel
Buttrey
Vadim
Kutsyy
Nick
Ellis
Bill
Dunlap
Glenn
Treacy
Bob
Byers
Terry
Therneau
The actual command
that worked was
> tapply (
get(infile)$Spd[good.s], as.factor(
get(infile)$Hour )[good.s],
mean )
If you use this, make sure that the [good.s] is outside
the function call to as.factor()
The same worked for
the length with an extra command to change the NA to 0
> x
<- tapply(
get(infile)$Spd[good.s], as.factor(
get(infile)$Hour )[good.s], length
)
> x[is.na(x)] <- 0
Thanks again. The responses were really fast and
useful.
Win
Lambert |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Existence of Methods, Prof Brian Ripley |
|---|---|
| Next by Date: | recommendation for a PC system, Chi-tsung Wu |
| Previous by Thread: | Existence of Methods, JRoyRobertson |
| Next by Thread: | recommendation for a PC system, Chi-tsung Wu |
| Indexes: | [Date] [Thread] [Top] [All Lists] |