========================================================
8 May 2000
========================================================

> library(Hmisc,T)
> page(cut2)
> x <- 3:7
> range(x)
[1] 3 7
> x <- 3:7
> range(x)
> ?mean
> help(mean)
> args(mean)
> mean
> mean
function(x, trim = 0, na.rm = F)
{
	if(na.rm) {
		wnas <- which.na(x)
		if(length(wnas))
			x <- x[ - wnas]
	}
	if(mode(x) == "complex") {
		if(trim > 0)
			stop("trimming not allowed for c
omplex data"
				)
		return(sum(x)/length(x))
	}
	x <- as.double(x)
	if(trim > 0) {
		if(trim >= 0.5)
			return(median(x, na.rm = F))
		if(!na.rm && length(which.na(x)))
			return(NA)
		n <- length(x)
		i1 <- floor(trim * n) + 1
		i2 <- n - i1 + 1
		x <- sort(x, unique(c(i1, i2)))[i1:i2]
	}
	sum(x)/length(x)
}
> x
[1] 3 4 5 6 7
> x <- c(1,2,3,NA)
> x
[1]  1  2  3 NA
> mean(x)
[1] NA
> args(mean)
function(x, trim = 0, na.rm = F)
NULL
> mean(x, na.rm=T)
[1] 2
> mean(x,, T)
[1] 2
> mean(x=x)
[1] NA
> mean(x=x,na.rm=T)
[1] 2
> mean(x,na.rm=T)
[1] 2
> mean(na.rm=T, x=x)
[1] 2
> length(x)
[1] 4
> is.na(x)
[1] F F F T
> x
[1]  1  2  3 NA
> mean(is.na(x))
[1] 0.25
> mean(!is.na(x))
[1] 0.75
> 1*is.na(x)
[1] 0 0 0 1
> table(is.na(x))
 FALSE TRUE 
     3    1
> x
[1]  1  2  3 NA
> x[3]
[1] 3
> x[-3]
[1]  1  2 NA
> x[3:4]
[1]  3 NA
> x[c(1,4)]
[1]  1 NA
> x > 2
[1]  F  F  T NA
> x[x > 2]
[1]  3 NA
> x > 2
[1]  F  F  T NA
> x > 2 & !is.na(x)
[1] F F T F
> x[x > 2 & !is.na(x)]
[1] 3
> is.inf(x)
[1] F F F F
> is.nan(x)
[1] F F F F
> is.character(x)
[1] F
> is.numeric(x)
[1] T
> is.double(x)
[1] T
> is.integer(x)
[1] F
> is.integer(1:3)
[1] T
> is.integer(c(1:3,NA))
[1] T
> x
[1]  1  2  3 NA
> is.integer(x)
[1] F
> x[2:4]
[1]  2  3 NA
> w <- c(cat=1, dog=2, giraffe=3)
> w
 cat dog giraffe 
   1   2       3
> w['cat']
 cat 
   1
> w[c('cat','giraffe')]
 cat giraffe 
   1       3
> #  data new; keep w; where name='cat' OR name='giraffe';
> w[c('cat','graffe')]
 cat graffe 
   1     NA
> x
[1]  1  2  3 NA
> sex <- c('f','m','f','f')
> sex
[1] "f" "m" "f" "f"
> x[sex=='f']
[1]  1  3 NA
> sex[is.na(x)]
[1] "f"
> sex[x < 2]
[1] "f" "" 
> sex
[1] "f" "m" "f" "f"
> x
[1]  1  2  3 NA
> X <- cbind(a=1:3, b=3:5)
> X
     a b 
[1,] 1 3
[2,] 2 4
[3,] 3 5
> X[c(1,3),]
     a b 
[1,] 1 3
[2,] 3 5
> X[, c('a','b')]
     a b 
[1,] 1 3
[2,] 2 4
[3,] 3 5
> X[, 1:2]
     a b 
[1,] 1 3
[2,] 2 4
[3,] 3 5
> X[, 1]
[1] 1 2 3
> X[, 1,drop=F]
     a 
[1,] 1
[2,] 2
[3,] 3
> X[, 'b']
[1] 3 4 5
> X[1:2, 'b']
[1] 3 4
> Z <- list(x=c(1,3), y=c('cat','dog','giraffe'))
> Z
$x:
[1] 1 3

$y:
[1] "cat"     "dog"     "giraffe"

> Z$x
[1] 1 3

> Z$y
[1] "cat"     "dog"     "giraffe"
> Z[[2]]
[1] "cat"     "dog"     "giraffe"
> Z[2]
$y:
[1] "cat"     "dog"     "giraffe"

> Z[['y']]
[1] "cat"     "dog"     "giraffe"
> Z$y
[1] "cat"     "dog"     "giraffe"
> Z[2:1]
$y:
[1] "cat"     "dog"     "giraffe"

$x:
[1] 1 3

> table(Z$y)
 cat dog giraffe 
   1   1       1
> mean(Z$x)
[1] 2
> Z[[1:2]]
[1] 3
> Z
$x:
[1] 1 3

$y:
[1] "cat"     "dog"     "giraffe"

> Z[[2:2]]
[1] "cat"     "dog"     "giraffe"
> Z[[2:3]]
[1] "giraffe"
> 1:2
[1] 1 2
> c(1,2)
[1] 1 2
> 2:2
[1] 2
> Z
$x:
[1] 1 3

$y:
[1] "cat"     "dog"     "giraffe"

> as.data.frame(Z)
Error in data.frameAux.list(x, na.stri..: arguments impl
y differing number of rows: 2, 3
> g <- data.frame(x=1:3, sex=c('female','male','male'))
> g
  x    sex 
1 1 female
2 2   male
3 3   male
> g$x
[1] 1 2 3
> g$sex
[1] female male   male  
> class(Z)
NULL
> class(g)
[1] "data.frame"
> print.data.frame(g)
  x    sex 
1 1 female
2 2   male
3 3   male
> print(g)
  x    sex 
1 1 female
2 2   male
3 3   male
> g
  x    sex 
1 1 female
2 2   male
3 3   male
> g
  x    sex 
1 1 female
2 2   male
3 3   male
> g[1:2,]
  x    sex 
1 1 female
2 2   male
> g[1:2,'sex']
[1] female male  
> g[3:1,]
  x    sex 
3 3   male
2 2   male
1 1 female
> attributes(g)
$names:
[1] "x"   "sex"

$row.names:
[1] "1" "2" "3"

$class:
[1] "data.frame"

> attributes(Z)
$names:
[1] "x" "y"

> attr(Z,'date') <- '8May00'
> Z
$x:
[1] 1 3

$y:
[1] "cat"     "dog"     "giraffe"

attr(, "date"):
[1] "8May00"
> is.list(Z)
[1] T
> is.list(g)
[1] T
> is.data.frame(g)
[1] T
> is.data.frame(Z)
[1] F
> class(Z)
NULL
> class(g)
[1] "data.frame"
> plot(g)
> args(assign)
function(x, value, frame, where = NULL, ...)
NULL
> g
  x    sex 
1 1 female
2 2   male
3 3   male
> g$sex
[1] female male   male  
> attributes(g$sex)
$levels:
[1] "female" "male"  

$class:
[1] "factor"

> as.integer(g$sex)
[1] 1 2 2
> g$sex == 'male'
[1] F T T
> g
  x    sex 
1 1 female
2 2   male
3 3   male
> g[sex=='male',]
[1] x   sex
< 0 rows> 
> sex
[1] "f" "m" "f" "f"
> g[g$sex=='male',]
  x  sex 
2 2 male
3 3 male
> a <- c(1,1,2,2)
> a
[1] 1 1 2 2
> a <- c(0,0,1,1)
> a
[1] 0 0 1 1
> b <- factor(a, 0:1, c('bad','good'))
> levels(b)
[1] "bad"  "good"
> b
[1] bad  bad  good good
> b <- factor(a, 1:0, c('good','bad'))
> b
[1] bad  bad  good good
> as.numeric(b)
[1] 2 2 1 1
> b <- factor(a, 1:0, c('1:good','0:bad'))
> b
[1] 0:bad  0:bad  1:good 1:good
> attributes(b)
$levels:
[1] "1:good" "0:bad" 

$class:
[1] "factor"

> args(factor)
function(x, levels = sort(unique(x), na.last = T), 
	labels = as.character(levels), exclude = NA)
NULL
> library(rpart)
> help(library='Hmisc')
> .First <- function() invisible(library(Hmisc,T))
> find(.First)
[1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
> help(library='rpart')
> ?cut
> ?cut2
> hospital <- cleanup.import(hospital)
> attach(hospital, pos=1, use.names=F)
> sex <- factor(sex, 1:2, c('male','female'))
> antibiotic <- factor(antibiotic, 1:2,
+ c('yes','no'))
> bculture <- factor(bculture, 1:2,
+   c('yes','no'))
> service <- factor(service, 1:2, 
+  c('medical','surgical'))
> label(temp) <- 'Temperature'
> search()
 [1] "hospital"                                         
 [2] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"   
 [3] "C:\\Program Files\\sp2000\\library\\Hmisc\\_Data" 
. . . .
> detach(1, 'hospital2')
> search()[1:3]
[1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"  
[2] "C:\\Program Files\\sp2000\\library\\Hmisc\\_Data"
[3] "C:\\Program Files\\sp2000\\splus\\_Functio"      
> label(hospital2$temp)
[1] "Temperature"
> page(describe(hospital2),multi=T)
> page(summary(hospital2))
> datadensity(hospital2)
> datadensity(hospital2,
+  group=hospital2$sex,
+  col=1:2)
> datadensity(hospital2,
+  group=hospital2$sex,
+  col=c(1,5))
> levels(hospital2$sex)
[1] "male"   "female"
> #x[x==99] <- NA
> #x[x==99] <- NA
> hist.data.frame(hospital2)
> ecdf(hospital2)
> ecdf(hospital2, group=
+  hospital2$sex, col=c(1,5))
> plot(varclus(~., data=
+  hospital2))
> hospital <- hospital2
> #remove('hospital2')
> search()
 [1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
 [2] "C:\\Program Files\\sp2000\\library\\Hmisc\\_Data"
. . . .
> objects(1)
 [1] ".First"       ".Last.value" 
 [3] ".Random.seed" "X"           
 [5] "Z"            "a"           
 [7] "b"            "g"           
 [9] "hospital"     "last.dump"   
[11] "last.warning" "sex"         
[13] "w"            "x"           
[15] "y"           
> find(hospital)
[1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
> find(age)
Error: Object "age" not found
> attach(hospital)
> search()
 [1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
 [2] "hospital"                    
 [3] "C:\\Program Files\\sp2000\\library\\Hmisc\\_Data"
. . . .
> 
> find(age)
[1] "hospital"
> find(a)
[1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
> find(ffff)
Error: Object "ffff" not found
> sex <- 'male'
> age <- c(10,20)
> find(age)
[1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
[2] "hospital"                     
> age
[1] 10 20
> hospital$age
 [1] 30 73 40 47 25 82 60 56 43 50
[11] 59  4 22 33 20 32 36 69 47 22
[21] 11 19 67 43 41
> remove('age')
> find(age)
[1] "hospital"

> masked()
[1] ".Random.seed" "last.warning"
[3] "sex"         
> # hospital$newvar <- NULL
> data.dump(c('hospital','a'),
+  '/windows/temp/dumpdata.sdd')
[1] "/windows/temp/dumpdata.sdd"
> data.dump(objects(1),
+  '/windows/temp/dumpdata.sdd')
[1] "/windows/temp/dumpdata.sdd"
> sort(hospital$age)
 [1]  4 11 19 20 22 22 25 30 32 33
[11] 36 40 41 43 43 47 47 50 56 59
[21] 60 67 69 73 82
> ?order
> i <- order(hospital$sex, hospital$age)
> i
 [1] 12 21 22 20 16 17 19 10  7 18
[11]  6 15 13  5  1 14  3 25  9 24
[21]  4  8 11 23  2
> page(hospital[i,])
> i <- order(hospital$sex, -hospital$age)
> page(hospital[i,])
> objects()
 [1] ".First"       ".Last.value" 
 [3] ".Random.seed" "X"           
 [5] "Z"            "a"           
 [7] "b"            "g"           
 [9] "hospital"     "i"           
[11] "last.dump"    "last.warning"
[13] "sex"          "w"           
[15] "x"            "y"           
> remove('sex',where=1)
Warning messages:
  "sex" to be removed not in databa
se in: remove("sex", where = 1)
> sex
 [1] female female female female
 [5] female male   male   female
 [9] female male   female male  
[13] female female female male  
[17] male   male   male   male  
[21] male   male   female female
[25] female
> search()
 [1] "C:\\Program Files\\sp2000\\us
ers\\Splus\\_Data"
 [2] "hospital"                    
. . . .
> find(age)
[1] "hospital"
> tapply(age, sex, mean, na.rm=T)
     male   female 
 39.27273 42.78571
> tapply(age, sex, median, na.rm=T)
 male female 
   36     42
> tapply(age, sex, quantile, na.rm=T)
$male:
   0%  25%  50%  75% 100% 
    4 20.5   36   55   82

$female:
   0%   25%  50%   75% 100% 
   20 30.75   42 53.75   73

> w <- tapply(age, sex, quantile, na.rm=T)
> w$age
NULL
> w$female
   0%   25%  50%   75% 100% 
   20 30.75   42 53.75   73
> tapply(age, list(sex,service), mean, na.rm=T)
        medical surgical 
  male 53.50000 36.11111
female 52.71429 32.85714
> by(age, list(sex), describe)
:male
x[i,  , drop = F] 

 1  Variables      11  Observations
-----------------------------------
-----------------------------------
-----
x 
  n missing unique  Mean  .05 
 11 0       11     39.27  7.5
  .10  .25  .50  .75  .90  .95 
 11.0 20.5 36.0 55.0 69.0 75.5

           4 11 19 22 32 36 47 50 
Frequency  1  1  1  1  1  1  1  1
        %  9  9  9  9  9  9  9  9
          60 69 82 
Frequency  1  1  1
        %  9  9  9
-----------------------------------
-----------------------------------
-----
--------------------------
:female
x[i,  , drop = F] 

 1  Variables      14  Observations
-----------------------------------
-----------------------------------
-----
x 
  n missing unique  Mean   .05 
 14 0       13     42.79 21.30
   .10   .25   .50   .75   .90 
 22.90 30.75 42.00 53.75 64.60
   .95 
 69.10

          20 22 25 30 33 40 41 43 
Frequency  1  1  1  1  1  1  1  2
        %  7  7  7  7  7  7  7 14
          47 56 59 67 73 
Frequency  1  1  1  1  1
        %  7  7  7  7  7
-----------------------------------
-----------------------------------
-----
> describe(hospital[c('age','temp')])
hospital[c("age", "temp")] 

 2  Variables      25  Observations
-----------------------------------
-----------------------------------
-----
age 
  n missing unique  Mean  .05 
 25 0       22     41.24 12.6
  .10  .25  .50  .75  .90  .95 
 19.4 25.0 41.0 56.0 68.2 72.2

lowest :  4 11 19 20 22
highest: 60 67 69 73 82 
-----------------------------------
-----------------------------------
-----
temp : Temperature 
  n missing unique  Mean   .05 
 25 0       12     98.31 97.12
   .10   .25   .50   .75   .90 
 97.60 98.00 98.20 98.60 99.12
   .95 
 99.44

          96.8 97.0 97.6 97.8 
Frequency    1    1    2    1
        %    4    4    8    4
          98.0 98.2 98.4 98.5 
Frequency    5    3    2    1
        %   20   12    8    4
          98.6 99.0 99.2 99.5 
Frequency    3    3    1    2
        %   12   12    4    8
-----------------------------------
-----------------------------------
-----
> by(hospital[c('age','temp')],
+ list(sex,service), describe)
sex:male
service:medical
x[i,  , drop = F] 

 2  Variables      2  Observations
-----------------------------------
-----------------------------------
-----
age 
 n missing unique Mean 
 2 0       2      53.5

47 (1, 50%), 60 (1, 50%) 
-----------------------------------
-----------------------------------
-----
temp : Temperature 
 n missing unique  Mean 
 2 0       2      98.25

97.0 (1, 50%), 99.5 (1, 50%) 
-----------------------------------
-----------------------------------
-----
--------------------------
sex:female
service:medical
x[i,  , drop = F] 

 2  Variables      7  Observations
-----------------------------------
-----------------------------------
-----
age 
 n missing unique  Mean 
 7 0       7      52.71

          30 41 43 56 59 67 73 
Frequency  1  1  1  1  1  1  1
        % 14 14 14 14 14 14 14
-----------------------------------
-----------------------------------
-----
temp : Temperature 
 n missing unique  Mean 
 7 0       4      98.11

97.6 (2, 29%), 98.0 (3, 43%) 
98.6 (1, 14%), 99.0 (1, 14%) 
-----------------------------------
-----------------------------------
-----
--------------------------
sex:male
service:surgical
x[i,  , drop = F] 

 2  Variables      9  Observations
-----------------------------------
-----------------------------------
-----
age 
 n missing unique  Mean 
 9 0       9      36.11

           4 11 19 22 32 36 50 69 
Frequency  1  1  1  1  1  1  1  1
        % 11 11 11 11 11 11 11 11
          82 
Frequency  1
        % 11
-----------------------------------
-----------------------------------
-----
temp : Temperature 
 n missing unique Mean 
 9 0       7      98.2

          96.8 97.8 98.0 98.2 
Frequency    1    1    2    2
        %   11   11   22   22
          98.6 99.0 99.2 
Frequency    1    1    1
        %   11   11   11
-----------------------------------
-----------------------------------
-----
--------------------------
sex:female
service:surgical
x[i,  , drop = F] 

 2  Variables      7  Observations
-----------------------------------
-----------------------------------
-----
age 
 n missing unique  Mean 
 7 0       7      32.86

          20 22 25 33 40 43 47 
Frequency  1  1  1  1  1  1  1
        % 14 14 14 14 14 14 14
-----------------------------------
-----------------------------------
-----
temp : Temperature 
 n missing unique  Mean 
 7 0       6      98.66

          98.2 98.4 98.5 98.6 
Frequency    1    2    1    1
        %   14   29   14   14
          99.0 99.5 
Frequency    1    1
        %   14   14
-----------------------------------
-----------------------------------
-----
> 
> aggregate(hospital[c('age','temp')],
+  service, mean)
          service      age 
 medical  medical 52.88889
surgical surgical 34.68750
             temp 
 medical 98.14444
surgical 98.40000
> page(combine.levels)
> expand.grid(sex=levels(sex),
+   service=levels(service))
     sex  service 
1   male  medical
2 female  medical
3   male surgical
4 female surgical
> expand.grid(sex=levels(sex),
+ age=c(10,20,30))
     sex age 
1   male  10
2 female  10
3   male  20
4 female  20
5   male  30
6 female  30
> set.seed(131)
> x <- runif(1000)
> table(cut2(x,g=5))
 [0.00106,0.19995) 
               200
 [0.19995,0.38568) 
               200
 [0.38568,0.61303) 
               200
 [0.61303,0.81991) 
               200
 [0.81991,0.99941] 
               200
> levels(cut2(x,g=5))
[1] "[0.00106,0.19995)"
[2] "[0.19995,0.38568)"
[3] "[0.38568,0.61303)"
[4] "[0.61303,0.81991)"
[5] "[0.81991,0.99941]"
> 
> Cs(a)
[1] "a"
> Cs(c(a,cat))
[1] "c(a, cat)"
> Cs(a,cat)
[1] "a"   "cat"
> ?merge.levels
> w <- score.binary(wbc>median(wbc),
+  temp>median(temp))
> table(w)
 none wbc > median(wbc) 
   11                 2
 temp > median(temp) 
                  12
> w <- score.binary(wbc>median(wbc),
+  temp>median(temp),
+  wbc > median(wbc) & temp>median(temp))
> table(w)
 none wbc > median(wbc) 
   11                 2
 temp > median(temp) 
                   3
 wbc > median(wbc) & temp > median(
	temp) 
                                   
      9
> v <- interaction(
+  cut2(wbc,g=2), cut2(temp,g=2))
> table(v)
 [ 3, 6).[96.8,98.2) 
                   5
 [ 6,14].[96.8,98.2) 
                   5
 [ 3, 6).[98.2,99.5] 
                   2
 [ 6,14].[98.2,99.5] 
                  13
> v <- interaction(
+  cut2(wbc,g=2), cut2(temp,g=2),
+  sep=' & ')
> table(v)
 [ 3, 6) & [96.8,98.2) 
                     5
 [ 6,14] & [96.8,98.2) 
                     5
 [ 3, 6) & [98.2,99.5] 
                     2
 [ 6,14] & [98.2,99.5] 
                    13

============================================
9 May 2000
============================================
		Command History from May 9, 2000 - A.M.

		IMPUTATION
		
age <- c(1,2,NA,4)
age.i <- impute(age)   ### default is median
age.i
attributes(age.i)
is.imputed(age)
is.imputed(age.i)
mean(age.i)
mean(age.i[!is.imputed(age.i)])

?transcan   #Has Details on Multiple Imputation

		BOOTSTRAPPING
		
sample(1:4,r,replace=T)    ####  with replacement
sample(1:4,4,replace=T)    ####  with replacement
sample(1:4,4,replace=T)    ####  with replacement
age[sample(1:4,4,replace=T)]
set.seed(1)
x <-runif(100)
b <- bootstrap(x,median, B=500)
?bootstrap
limits.emp(b)
b <- bootstrap(x,mean, B=500)
limits.emp(b)
b
?bpower.sim   ###FH likes to use for simulation - very fast
bpower.sim( .1, .2, n1=500, n2=300, nsim=10000)     ####  10,000 is default
bpower.sim( .1, .2, n1=500, n2=300, nsim=100000)     ####  10,000 is default
args(bpower.sim)

binconf(0,100)

		CODE FROM ?BPOWER
		
# Plot power vs. n for various odds ratios  (base prob.=.1)
n  <- seq(10, 1000, by=10)
OR <- seq(.2,.9,by=.1)
plot(0, 0, xlim=range(n), ylim=c(0,1), xlab="n", ylab="Power", type="n")
for(or in OR) {
  lines(n, bpower(.1, odds.ratio=or, n=n))
  text(350, bpower(.1, odds.ratio=or, n=350)-.02, format(or))
}
# Another way to plot the same curves, but letting labcurve do the
# work, including labeling each curve at points of maximum separation
pow <- lapply(OR, function(or,n)list(x=n,y=bpower(p1=.1,odds.ratio=or,n=n)),
              n=n)
names(pow) <- format(OR)
labcurve(pow, pl=T, xlab='n', ylab='Power')
?spower

		UNIFIED NON-PARAMETRICS
args(spearman2)
s <- spearman2(duration~ age+sex+temp, data=hospital)
s
s <- spearman2(duration~ age+sex+temp, data=hospital, p=2)
s
plot(s)

		SUMMARY.FORMULA
		
?summary.formula
page(describe(titanic3))
s <-summary( survived ~ age + sex + parch + pclass, data=titanic3)
options(digits=3)
###########   Variables analyzed separately
page(s)
class(s)
plot(s)

s2 <-summary( age ~ sex+pclass, data=titanic3)
plot(s2)   ### Mean age stratified

s2 <-summary( age ~ sex+pclass, data=titanic3, fun=quantile)
s2

	####  Using your own f/n
g<-function(y)  c(Mean=mean(y),Median=median(y))

s2 <-summary( age ~ sex+pclass, data=titanic3, fun= g )
s2
	#####  Plot(s2) would plot only first element
plot(s2, which=1:2, pch=1:2)  	### Solution to above

	#####  Cross-Classifications in SUMMARY
s3 <-summary( age ~ sex+pclass, data=titanic3, fun= g , method='cross' )
page(s3)
#####   No plot method available

	#####   REVERSE Tables
sr <-summary( sex ~ age +pclass, data=titanic3, method='reverse' ) 
page(sr)
plot(sr)


		ASSIGNMENT 4.3
s <-summary( wbc ~ age + sex + antibiotic + bculture +service , g=3, data=hospital)
s
s <-summary( wbc ~ age + sex + antibiotic + bculture +service , fun=median , g=3, data=hospital)
s
plot(s)

result.4d <-summary( wbc ~ age + sex + antibiotic + bculture +service , fun=g , g=3, data=hospital)
result.4d
plot(result.4d, which=1:2, pch=1:2, main='Mean and Median of WBS')
plot(result.4d, which=1:2, pch=1:2,xlab='Circle:mean Triangle:median', main='Mean and Median of WBS')

result.4a <- summary(temp ~ age + sex, fun = median, method='cross', data=hospital)
result.4a
summary( sex ~ age +pclass, data=titanic3, method='reverse' ) ### Reverse table

		####   Assignment 4.6
attach(hospital)
plot(temp, wbc, xlab='Temperature', main='White Blood Count vs Temperature')
text ( locator(1) , 'Comment')


===========================================================
Afternoon
===========================================================
> datadensity(titanic3)
> ecdf(titanic3)
> attach(titanic3)
> ecdf(age)
> ecdf(age,q=c(.25,.5))
> ecdf(age,group=sex,q=c(.25,.5))
> ecdf(age,group=sex,q=c(.25,.5),
+   col=c(1,3))
> ecdf(titanic3, group=titanic3$sex)

> par(mfrow=c(2,2))
> ecdf(titanic3, group=titanic3$sex)
> datadensity(diabetes)
> par(mfrow=c(1,1))
> datadensity(diabetes)
> ecdf(diabetes, group=diabetes$gender,
+  label.curve=F, col=1:2)
> ecdf(diabetes$chol)
> ecdf(diabetes$chol,
+ datadensity='rug')
> ecdf(diabetes$chol,
+ datadensity='hist')
> ecdf(diabetes$chol,
+ datadensity='density')

> bwplot(~chol, data=diabetes)
> bwplot(gender~chol, data=diabetes)
> bwplot(gender~chol | frame, data=diabetes)
> bwplot(gender~chol | frame*location, data=diabetes)
> bwplot(gender~chol | frame*location, data=diabetes,
+  panel=panel.bpplot)
> ?panel.bpplot
> set.seed(13)
> x <- rnorm(1000)
> g <- sample(1:6, 1000, replace=T)
> x[g==1][1:20] <- rnorm(20)+3   # contaminate 20 x's for group 1
> 
> # default trellis box plot
> bwplot(g ~ x)
> 
> # box-percentile plot with data density (rug plot)
> bwplot(g ~ x, panel=panel.bpplot, probs=seq(.01,.49,by=.01), datadensity=T)
> # add ,scat1d.opts=list(tfrac=1) to make all tick marks the same size
> # when a group has > 125 observations
> 
> # small dot for means, show only .05,.125,.25,.375,.625,.75,.875,.95 quantiles
> bwplot(g ~ x, panel=panel.bpplot, cex=.3)
> 
> bwplot(g ~ x, panel=panel.bpplot, probs=c(.025,.1,.25), means=F, qref=F)
> 
> # continuous plot up until quartiles ("Tootsie Roll plot")
> bwplot(g ~ x, panel=panel.bpplot, probs=seq(.01,.25,by=.01))
> 
> bwplot(g ~ x, panel=panel.bpplot, probs=seq(.25,.49,by=.01))
> 
> bwplot(g ~ x, panel=panel.bpplot, probs=c(.025,seq(.25,.49,by=.01)))
> 
> bwplot(g ~ x, panel=panel.bpplot, probs=c(.1,.2,.3,.4), qref=c(.5,.2,.8))
> 
> bwplot(g ~ x, panel=panel.bpplot, nout=.05, scat1d.opts=list(frac=.01))
> 
> bwplot(g ~ x, panel=panel.bpplot, nout=5)
> 
> bwplot(~x , panel=panel.bpplot, probs=seq(.00,.5,by=.001), 
+        datadensity=T, scat1d.opt=list(preserve=T))
> 
> search()
 [1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
 [2] "titanic3"                  
. . . .
> plsmo(age, survived, 
+  datadensity=T)
> plsmo(age, survived, 
+  group=sex, col=1:2,datadensity=T)
> plsmo(age, survived, 
+  group=pclass, col=1:3,datadensity=T)
> masked()
[1] ".Random.seed" "last.warning"
[3] "s"           
> plsmo(age, survived, 
+  group=interaction(pclass,sex),
+  col=1:6, datadensity=T)
> plsmo(age, survived, 
+  group=interaction(pclass,sex),
+  col=1:6, datadensity=T, lty=1)
> g <- crosstabs(~ gender + frame,
+ data=diabetes)
Error in na.fail.data.fram..: mis
sing values not allowed: found in
 frame

> w <- summarize(age,
+   llist(sex,pclass),
+   function(y)sum(!is.na(y)))
> w
     sex pclass age 
1 female    1st 133
2 female    2nd 103
3 female    3rd 152
4   male    1st 151
5   male    2nd 158
6   male    3rd 349
> attributes(w)
$names:
[1] "sex"    "pclass" "age"   

$class:
[1] "data.frame"

$row.names:
[1] 1 2 3 4 5 6

> w$age
Age 
[1] 133 103 152 151 158 349
> w$pclass
Passenger Class 
[1] 1st 2nd 3rd 1st 2nd 3rd
> w$sex
Sex 
[1] female female female male  
[5] male   male  
> g <- naclus(diabetes)
> plot(g)
> naplot(g)
> v <- varclus(~.,data=diabetes)
> plot(v)

> s <- summary(gender ~ age +
+  chol + ratio + frame + glyhb +
+  location, data=diabetes,
+  method='reverse')
> plot(s)
> search()
 [1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
 [2] "titanic3"                  
 [3] "C:\\Program Files\\sp2000\\library\\Hmisc\\_Data"
. . . . 
> detach(2)
> attach(diabetes)
> g <- summarize(glyhb,
+  llist(age=cut2(age,g=5),
+        gender, frame),
+  median, na.rm=T)
> dotplot(gender ~ glyhb |
+  age*frame, data=g)
> dotplot(age ~ glyhb |
+  gender*frame, data=g)
> 
> g <- summarize(glyhb,
+  llist(gender,frame),
+  quantile, probs=c(.25,.5,.75),
+  na.rm=T)
> ?summarize
> g <- summarize(glyhb,
+  llist(gender,frame),
+  smedian.hilow, conf.int=.5)
> Dotplot(frame ~ 
+  Cbind(glyhb,Lower,Upper) |
+  gender, data=g)
> ?Dotplot
> names(g)
[1] "gender" "frame"  "glyhb" 
[4] "Lower"  "Upper" 
> 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)
> 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)
> 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)
> Dotplot(month ~ Cbind(y, Lower, Upper) | year, data=s) 
> xYplot(Cbind(y,Lower,Upper) ~ month, groups=year, data=s, 
+        keys='lines', method='alt')
> xYplot(Cbind(y,Lower,Upper) ~ month, groups=year, data=s, 
+        keys='lines', method='bands')
> xYplot(Cbind(y,Lower,Upper) ~ month, groups=year, data=s, 
+        keys='lines', method='bars')
> s <- summarize(y, llist(month,year), smean.cl.boot)
> xYplot(Cbind(y, Lower, Upper) ~ month | year, data=s)
> xYplot(Cbind(y, Lower, Upper) ~ 
+  month, groups=year, data=s)
> search()
 [1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
 [2] "dfr"                           
. . . .
> detach(2)
> search()
 [1] "C:\\Program Files\\sp2000\\users\\Splus\\_Data"
 [2] "diabetes"                      
. . . .

> f <- lm(glyhb ~ age + gender)
Error in function(object, ....: missi
ng values not allowed: found in glyhb
> f <- lm(glyhb ~ age + gender,
+   na.action=na.delete)
> r <- resid(f)
> length(r)
[1] 403
> length(age)
[1] 403
> plot(age, r)
Warning in axes: No room for ylab
> xYplot(r ~ age)
> xYplot(r ~ age, type='p')
> xYplot(r ~ age, method='quantile')
> args(xYplot)
function(formula, data = sys.parent(
	1), groups = NULL, prepanel
	 = prepanel.xYplot, panel = 
	"panel.xYplot", ..., ylab = 
	attr(eval(formula[[2]], data
	), "label"), subset = T)
NULL
>