dos('dir')
Age <- c(6, 5, 4, 8, 10, 5)
Age
mean(Age)
quantile(Age, c(.25,.75))
cor(Age, Age)
1:100
1:20
1:20
1:20
seq(1,20)
resize()
?cut
help(cut)
args(mean)
names(mean)
mean(c(1,2,NA))
mean(c(1,2,NA),na.rm=T)
Age
age2 <- c(Age,NA)
age2
mean
x <- c(3,6,9,10,2.2,NA,NA,6.7)
x[1:3]
x[c(1,2,5)]
x[-(1:3)]
x
is.na(x)
x[!is.na(x)]
x
y <- 1:9
list(x=x, y=y)
data.frame(x,y)
x
y <- 1:8
data.frame(x,y)
d <- data.frame(x,y)
d
d$x
d$y
d[[1]]
d[,1]
d[1]
d[1:2]
d[1:5,1:2]
d[1:5,]
library(Hmisc,T)
help(library='Hmisc')
find(cut2)
find(cut)
masked()
search()
masked(2)
find(label)
label
desc <- describe(hospital)
page(desc, multi=T)
label(hospital$duration) <- 'Duration of hospital stay'
hospital$sex <- factor(hospital$sex,
  1:2, c('male','female'))
label(hospital$temp) <- 'First temperature following admission'
hospital$antib <- factor(hospital$antib, 1:2, c('yes','no'))
hospital$bact <- factor(hospital$bact, 1:2, c('yes','no'))
hospital$bact
names(hospital)
table(hospital$bculture)
hospital$bculture <- factor(hospital$bculture,
 1:2, c('yes','no'))
hospital$bact <- NULL
hospital$service <- factor(hospital$service,
1:2, c('med','surg'))
desc <- describe(hospital)
page(desc, multi=T)
datadensity(hospital)
source('a:/hospital.s')
dos('dir a:')
source('a:/hospital.q')
desc <- describe(hospital)
page(desc, multi=T)
datadensity(titanic3)
hist.data.frame(titanic3)
datadensity(titanic3)
ecdf(titanic3)
nac <- naclus(titanic3)
plot(nac)
naplot(nac)
page(describe(titanic3),multi=T)
library(rpart)
g <- rpart(is.na(age) ~
 survived + sex + pclass,
 data=titanic3)
plot(g);text(g)
levels(titanic3$pclass)
?rpart
?rpart.control
 survived + sex + pclass,
g <- rpart(is.na(age) ~
 survived + sex + pclass,
 data=titanic3,
 control=rpart.control(minsplit=8))
plot(g);text(g)
g <- rpart(is.na(age) ~
 survived + sex + pclass,
 data=titanic3,
 control=rpart.control(minsplit=6,cp=.05))
plot(g);text(g)
g <- rpart(is.na(age) ~
 survived + sex + pclass,
 control=rpart.control(
  minsplit=6,cp=.001),
 data=titanic3)
plot(g);text(g)
g$cptable
plot(titanic3)
find(pclass)
find(titanic3)
attach(titanic3)
find(pclass)
table(pclass)
table(titanic3$pclass)
search()
masked()
age <- 1:3
find(age)
plot(age, survived)
masked()
remove('age')
find(age)
page(search())
edit(names(hospital))
tapply(age,pclass,mean,
 na.rm=T)
tapply(age,pclass,var,na.rm=T)
tapply(age,pclass,quantile)
tapply(age,pclass,quantile,
 na.rm=T)
tapply(age, list(sex,pclass),
 mean, na.rm=T)
by(age, list(pclass=pclass),
 FUN=describe)
by(age, pclass,FUN=describe)
by(age, list(Passenger.Class=pclass),
 FUN=describe)
aggregate(titanic3[c('age','parch')],
 sex, FUN=mean, na.rm=T)
expand.grid(sex=levels(sex),
 pclass=levels(pclass))
x <- c('cat','dog','cat','giraffe')
x <- factor(x)
levels(x)
?merge.levels
levels(x) <- list(Domestic=c('cat','dog'),
   Wild='giraffe')
x
levels(x)
#  For help do ?merge.levels
levels(x) <- list(Home='Domestic',
   Other='Wild')
x
levels(x) <- list('Home Creature'='Home')
x
x <- c(1, 2, 2, 7, 10)
y <- c(0, 0, 0, 3, 20)
z <- score.binary(x>=7, y>=7)
z
levels(z)
z <- score.binary(x>=7, y>=7, x>=7 & y>=7)
z
find(z)
age <- c(1,2,NA,4)
age.i <- impute(age)
age.i
is.imputed(age.i)
attributes(age.i)
mean(age.i)
impute(age, 50)
impute(age, mean)
?impute
age.i
y <- 4:1
plot(x, y)
plot(age.i, y)
for(i in 1:2) {
 s <- is.imputed(age.i)
 if(i==1)plot(age.i[!s],y[!s],pch=1) else
 plot(age.i[s], y[s], pch=2)
}
for(i in 1:2) {
 s <- is.imputed(age.i)
 if(i==1)plot(age.i[!s],y[!s],pch=1) else
 points(age.i[s], y[s], pch=2)
}
  #    subset=!is.imputed(age.i)  
# Estimate var(sample median from sample
# of size 50 from log-normal distribution
n <- 50
reps <- 400
meds <- single(reps)
set.seed(171)
for(i in 1:reps) {
 x <- exp(rnorm(n))
 meds[i] <- median(x)
}
var(meds)
hist(meds, nclass=40)
hist(meds, nclass=30)
hist(meds, nclass=25)
qqnorm(meds)
qqline(meds)
single(5)
set.seed(1)
for(i in 1:reps) {
 x <- exp(rnorm(n))
 meds[i] <- median(x)
}
var(meds)
set.seed(173)
x <- matrix(exp(rnorm(n*reps), nrow=reps,
 byrow=T)
)
x <- matrix(exp(rnorm(n*reps)), nrow=reps,
 byrow=T)
dim(x)
meds <- apply(x, 1, median)
dim(meds)
length(meds)
1:4
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
sample(1:4, 4, rep=T)
x <- exp(rnorm(100))
hist(x, nclass=15)
mean(x)
median(x)
b <- bootstrap(x, mean, B=1000)
?bootstrap
summary(b)
plot(b)
?bootstrap


page(w, multi=T) 
resize() 
ecdf(wbc) 
library(Hmisc,T) 
ecdf(wbc) 
attach(hospital) 
ecdf(wbc) 
ecdf(wbc,group=sex) 
ecdf(hospital) 
datadensity(hospital, 
 group=hospital$sex, 
 col=1:2) 
?plsmo 
search()[1:3] 
detach(2) 
attach(titanic3) 
plsmo(age, survived, 
 datadensity=T) 
plsmo(age, survived, 
 group=sex, datadensity=T, 
 col=1:2) 
plsmo(age, survived, 
 group=interaction(sex,pclass), 
 datadensity=T, col=1:6) 
page(describe(diabetes)) 
datadensity(diabetes) 
nac <- naclus(diabetes) 
plot(nac) 
naplot(nac) 
names(diabetes) 
v <- varclus(~. , 
  data=diabetes[,-1]) 
plot(v) 
bwplot(cut2(age,g=5) ~  
  glyhb | sex*frame, 
  data=diabetes) 
bwplot(cut2(age,g=5) ~  
  glyhb | gender*frame, 
  data=diabetes) 
bwplot(cut2(age,g=4) ~  
  glyhb | frame, 
  data=diabetes) 
bwplot(cut2(age,g=4) ~  
  glyhb | frame, 
  data=diabetes, 
  panel=panel.bpplot) 
?panel.bpplot 
bwplot(cut2(age,g=4) ~  
  glyhb | frame, 
  data=diabetes, 
  panel=panel.bpplot, 
  probs=seq(.01,.49,by=.01)) 
f <- summary(glyhb ~ 
  gender + age + chol + 
  height + weight + hdl, 
  data=diabetes, 
  fun=smedian.hilow, 
  conf.int=.5) 
f <- summary(glyhb ~ 
  gender + age + chol + 
  gender + age + chol + 
f <- summary(glyhb ~ 
  gender + age + chol + 
  height + weight + hdl, 
  data=diabetes, 
  fun=function(y) 
   quantile(y,c(.25,.5,.75))) 
plot(f) 
plot(f, which=1:3, 
 pch=c('.','[',']') 
) 
plot(f, which=1:3, 
 pch=c('[','.',']') 
) 
ecdf(~chol | gender,data=diabetes) 
ecdf(~chol,groups=gender,data=diabetes) 
ecdf(~chol,groups=gender,data=diabetes, 
  q=c(.25,.5,.75)) 
ecdf(diabetes, 
 group=diabetes$gender, 
 label.curve=F) 
par(mfrow=c(2,3)) 
ecdf(diabetes, 
 group=diabetes$gender, 
 label.curve=F) 
par(mfrow=c(2,2)) 
ecdf(diabetes, 
 group=diabetes$gender, 
 label.curve=F) 
par(mfrow=c(1,1)) 
search()[1:4] 
detach(2) 
attach(diabetes) 
masked() 
s <- summarize(glyhb, 
  llist(age=cut2(age,g=5), 
        gender, frame), 
  median, na.rm=T) 
page(s) 
options(width=70) 
page(s) 
dotplot(age~glyhb|gender*frame) 
dotplot(age~glyhb| 
 gender*frame, data=s) 
Dotplot(age~glyhb|frame, 
 groups=gender, data=s) 
Key() 
?Dotplot 
set.seed(111) 
dfr <- expand.grid(month=1:12, year=c(1997,1998), reps=1:100) 
attach(dfr) 
y <- abs(month-6.5) + 2*runif(length(month)) + year-1997 
s <- summarize(y, llist(month,year), smedian.hilow, conf.int=.5) 
xYplot(Cbind(y,Lower,Upper) ~ month, groups=year, data=s,  
       keys='lines', method='alt') 
s <- summarize(y, llist(month,year), quantile, 
               probs=c(.5,.05,.25,.75,.95), type='matrix') 
Dotplot(month ~ Cbind(y) | year, data=s) 
?Dotplot 
?setTrellis 
Dotplot(month ~ Cbind(y) | year, data=s, 
 lwd=3) 
trellis.par.get 
trellis.par.get() 
?trellis.par.get 
?xYplot 
dfr <- expand.grid(month=1:12, continent=c('Europe','USA'),  
                   sex=c('female','male')) 
attach(dfr) 
set.seed(13) 
y <- month/10 + 1*(sex=='female') + 2*(continent=='Europe') +  
  runif(48,-.15,.15) 
lower <- y - runif(48,.05,.15) 
upper <- y + runif(48,.05,.15) 
xYplot(Cbind(y,lower,upper) ~ month,subset=sex=='male'&continent=='USA') 
xYplot(Cbind(y,lower,upper) ~ month|continent,subset=sex=='male') 
xYplot(Cbind(y,lower,upper) ~ month|continent,groups=sex); Key() 
xYplot(Cbind(y,lower,upper) ~ month,groups=sex,subset=continent=='Europe') 
xYplot(Cbind(y,lower,upper) ~ month,groups=sex,subset=continent=='Europe',keys='lines') 
# keys='lines' causes labcurve to draw a legend where the panel is most empty 
xYplot(Cbind(y,lower,upper) ~ month,groups=sex,subset=continent=='Europe',method='bands') 
xYplot(Cbind(y,lower,upper) ~ month,groups=sex,subset=continent=='Europe',method='upper') 
?setps