subsetting.txt

head(airquality)
attributes(airquality)

#selects obs with Temp>80 and selects variables Ozone and Temp
subset1 <-subset(airquality, Temp > 80, select = c(Ozone, Temp))

#selects obs with Day = 1. Selects all vars except Temp. Use "==" since "Day" is integer
subset2 <- subset(airquality, Day == 1, select = -Temp)

#selects all obs and variables FROM Ozone TO Wind
subset3 <- subset(airquality, select = Ozone:Wind)

#selects values of Ozone where corresponding Temp >80. This will return a vector consisting of the values of Ozone.

with1 <- with(airquality, subset(Ozone, Temp > 80))

#to keep the vars (i.e. utput is not a vector) you use
with2 <- with(airquality, subset(airquality, Temp > 80))

#using expression below with result in with3=with1
with3 <- with(airquality, subset(airquality$Ozone, Temp > 80))

## sometimes requiring a logical 'subset' argument is a nuisance
#documentation of state.x77 in "http://www.inside-r.org/r-doc/datasets/state.abb"

head(state.x77)
attributes(state.x77)

#get rownames which are the values of the first column

nm <- rownames(state.x77)
nm

#returns a TRUE and FALSE vector. True when nm (or rowname) that follows the pattern in "grep"
# "%in%" is a logical operator. grep function is selects values of nm that starts with M.
# value: if FALSE, a vector containing the (integer) indices of the matches determined by grep is returned. 
# value: if TRUE, a vector containing the matching elements themselves is returned.

start_with_M <- nm %in% grep("^M", nm, value = TRUE)
start_with_M

#selects values of state.x77 that is TRUE in "start_with_M". 
#The third argument (Illiteracy:Murder) specifies the variables to be selected.
#The pattern in selecting specific states that starts with M is:
#1) use %in% (logical operator) and grep (fcn) to create a TRUE and FALSE vector, TRUE if the state starts with M and FALSE if otherwise
#2) use subset to use the TRUE and FALSE vector to select all states are "TRUE"
#It's like getting the subset of state.x77 and start_with_M = TRUE

subset(state.x77, start_with_M, Illiteracy:Murder)

# but in recent versions of R this can simply be

start_with_M2<- subset(state.x77, grepl("^M", nm), Illiteracy:Murder)
start_with_M2

#whew!!!

from

http://stat.ethz.ch/R-manual/R-patched/library/base/html/subset.html

"Enjoy learning"