1.2 R R Windows, Macintosh, Linux(Unix) Windows Mac R Linux redhat, debian, vinelinux ( ) RjpWiki ( RjpWiki Wiki

Size: px

Start display at page:

Download "1.2 R R Windows, Macintosh, Linux(Unix) Windows Mac R Linux redhat, debian, vinelinux ( ) RjpWiki ( RjpWiki Wiki"

かずきたけすえ
5 years ago
Views:

1 R ( ) 1 R 1.1 R R R S-PLUS( ) S version 4( ) S (AT&T Richard A. Becker, John M. Chambers, and Allan R. Wilks ) S S R R S ( ) S GUI( ) ( ) R R R R yama/r/ R yamamoto@sm.u-tokai.ac.jp 1

2 1.2 R R Windows, Macintosh, Linux(Unix) Windows Mac R Linux redhat, debian, vinelinux ( ) RjpWiki ( RjpWiki Wiki yama/r/install.html R R > ( ) ( ) ( ) ( ) ( ) ( ) ( ) > R q() Windows 2.2 R (+, -, *, /, ^( )) (cos(), sin(), tan()) sqrt() log() pi(π = ) c() ( ) 10 mean() 2

3 > mean(c(148, 160, 159, 153, 151, 140, 158, 137, 149, 160)) [1] <- > height <- c(148, 160, 159, 153, 151, 140, 158, 137, 149, 160) > height [1] > mean(height) [1] ( ) M, F > sex <- c("f","m","m","f","f","f","m","f","m","f") ( ) tapply() > tapply(height, sex, mean) F M data.frame() > mydata <- data.frame(height,sex) > mydata height sex F M F $ mydata$height Excel R Excel 3

4 ( example1.dat) c:\tmp # (example1.dat) height weight sex "M" "M" "F" read.table exdata1 <- read.table(file="c:/tmp/example1.dat", header=t) read.table() file= ( ) \( ) /( ) ) read.table("c:/tmp/example1.dat") file= ("c:\\tmp\\example1.dat" ) R header=true (T ) $ exdata1$height attach ( ) height ls() rm() attach() exdata1 > ls() [1] "exdata1" "height" "mydata" "sex" > rm(height, sex) > attach(exdata1) detach() ( ) R mean(), median(), max(), min(), var(), sd() > mean(height) [1] 149 > var(height) [1] > sd(height) [1]

5 table() table() ( ) cut() > table(sex) sex F M > table(height) height > table(cut(height,seq(135,165,by=5))) (135,140] (140,145] (145,150] (150,155] (155,160] (160,165] seq() by= R > summary(exdata1) height weight sex Min. :137.0 Min. :29.00 F:14 1st Qu.: st Qu.:33.25 M:16 Median :149.0 Median :38.00 Mean :149.0 Mean : rd Qu.: rd Qu.:43.75 Max. :161.0 Max. :49.00 Min.( ) Max.( ) 1st Qu.( 1 ) 3rd Qu.( 3 ) Median( ) Mean( ) 2.4 hist() boxplot() hist() breaks= breaks= ( ) 0 5 breaks= > hist(height) > hist(height,breaks=3) > hist(height,br=c(136,140,144,148,152,156,160,170)) 5

6 hist() 1 par() par() > op <- par(mfrow=c(2,1)) # op > hist(height) > hist(height, br=6,col="lightblue", border="pink") > par(op) # > str(hist(height,plot=f)) List of 4 $ breaks : num [1:7] $ counts : int [1:6] $ intensities: num [1:6] $ mids : num [1:6] > h <- hist(height,plot=f) > h$counts [1] hist() stem(height) ( ) boxplot() ( ) ( 2.1 ) ( 2.1 ) > par(mfrow=c(1,2)) # > boxplot(height, main="height") > boxplot(height~sex, main="height~sex") height height~sex F M 2.1: ( ) ( ) 6

7 R plot() ( ) > plot(height) > plot(seq(1,length(height)),height,type="b",xlab="") # > plot(height,weight) # (factor text ) R > plot(exdata1$sex) > plot(exdata1) # > plot(weight~height, data=exdata1) # 2.5 R S travel separation 10 (bicycle.csv :LISP-STAT, 1997, ) separation travel lm > bicycle <- read.table("bicycle.csv", header=t, sep=",") #CSV ( ) sep="," > lm(separation ~ travel, data=bicycle) Call: lm(formula = separation ~ travel, data = bicycle) Coefficients: (Intercept) travel (Intercept) (travel) lm() lm() summary() 7

8 > bicycle.lm <- lm(separation ~ travel, data=bicycle) > summary(bicycle.lm) Call: lm(formula = separation ~ travel, data = bicycle) Residuals: Min 1Q Median 3Q Max Coefficients: Estimate Std. Error t value Pr(> t ) (Intercept) travel e-06 *** --- Signif. codes: 0 *** ** 0.01 * Residual standard error: on 8 degrees of freedom Multiple R-Squared: , Adjusted R-squared: F-statistic: on 1 and 8 degrees of freedom, p-value: 9.975e-006 > coef(bicycle.lm) (Intercept) travel R R R R 3.1 R demo( ) demo() R 2 help() apropos() help.search() ("") Windows GUI 8

9 R (numeric) (text) (logical) TRUE(T) FALSE(F) Inf( ) NAN(0 ) NA c() 2 1, sep() rep() > 1:5 [1] > seq(2,10,3) [1] > rep("f",3) [1] "F" "F" "F" > rep(1:3,c(2,3,4)) [1] mean, median, var, cor min, max, range (sum) (cumsum) (sort, rank, order) list() array() matrix() > in1 <- list(height,weight) > in2 <- array(c(height,weight), dim=c(30, 2)) > in3 <- matrix(c(height,weight), ncol=2) [[ ]] in1 9

10 > in1[[1]] [1] [20] > names(in1) <- c(" "," ") > in1$ [1] [20] ( %*%) (solve()) (eigen()) (svd()) (chol()) cov() eigen() > cov(in2) [,1] [,2] [1,] [2,] > eigen(cov(in2)) $values [1] $vectors [,1] [,2] [1,] [2,] R library() install.packages( ) library( ) 10

11 scan scan() # (example2.dat) ( ) file=" " sep=" " sep sep="," scan scan 2 10 > in4 <- scan("example2.dat") Read 60 items > in1 [1] [20] [39] [58] "" > in5 <- scan("example2.dat", list(0,0)) Read 30 lines Excel library(rodbc) ODBC Excel CSV CSV read.table() (SPSS SAS Minitab) foreign 3.5 R 11

12 3.5.1 R funcname <- function(arg1,..) expression funcname arg1,.. expression var > var function (x, y = x, na.rm = FALSE, use) { if (missing(use)) use <- if (na.rm) "complete.obs" else "all.obs" cov(x, y, use = use) } var cov ( ) tmean tmean <- function (x) { (sum(x)-max(x)-min(x))/(length(x)-2) } ( ) (tsum, tlen ) tmean2 <- function (x) { tsum <- sum(x)-max(x)-min(x) tlen <- length(x)-2 tsum/tlen } R (if, else, switch) (for, while, repeat) edit( ) edit() (<-, = ) source() R ( ) R for sum() for log(height) 12

13 > height-median(height) # [1] [20] > sum(abs(height-median(height))) # [1] 180 > sum((height-median(height))^2) # 2 [1] 1552 > (height-median(height))%*%(height-median(height)) # [,1] [1,] 1552 ( abs) (sum) R RjpWiki R 4 R R R GUI R 4.1 R Save workspace image? R.RData objects() ( S ls() ) pat=" " rm() > objects() [1] "height" "weight" "xname" > objects (pat="we*") #we [1] "weight" > rm("xname") sink() sink() 13

14 > sink("yama.log") > objects() > table(height) > mean(height) > sink() write() write.table() 4.2 R edit( ) Excel data.entry() de() R R Linux Emacs ESS Windows Meadow ESS Emacs Windwos WinEdt xyzz Tinn-R( ) R 4.3 R R GUI R (R Commandar) R R R R Rcmdr R SDI R R --sdi 4.4 R R RjpWiki Wiki R R The R Tips R 14

15 R The R Book Web RjpWiki R R R project RjpWiki R-Tips( ) R Tips( ) R tips( ) R ( ) 15

DAA09

DAA09 > summary(dat.lm1) Call: lm(formula = sales ~ price, data = dat) Residuals: Min 1Q Median 3Q Max -55.719-19.270 4.212 16.143 73.454 Coefficients: Estimate Std. Error t value Pr(> t ) (Intercept) 237.1326