Analizzare i dati con R - Ld17_code
Alessandra Santi
28 ottobre 2017
?R
R is a language and environment for statistical computing and graphics. It is a GNU project which is similar to the S language and environment which was developed at Bell Laboratories (formerly AT&T, now Lucent Technologies) by John Chambers and colleagues. R can be considered as a different implementation of S. There are some important differences, but much code written for S runs unaltered under R.
R provides a wide variety of statistical (linear and nonlinear modelling, classical statistical tests, time-series analysis, classification, clustering, …) and graphical techniques, and is highly extensible.
?RStudio
RStudio is an integrated development environment (IDE) for R. It includes a console, syntax-highlighting editor that supports direct code execution, as well as tools for plotting, history, debugging and workspace management.
RStudio is available in open source and commercial editions and runs on the desktop (Windows, Mac, and Linux) or in a browser connected to RStudio Server or RStudio Server Pro (Debian/Ubuntu, RedHat/CentOS, and SUSE Linux).
Packages and Help
Packages
- install.packages(‘dplyr’) Download and install
- library(dplyr) Load
- data(iris) Load a built-in dataset
Help
- ?mean Help function mean
- help(package = ‘dplyr’) Help package.
Working Directory
Where am I?
- getwd() Find the current working directory (where inputs are found and outputs are sent).
- setwd(‘C://file/path’) Change the current working directory.
Table of contents
- Vectors
Data Frames
- Libraries
- magrittr
- dplyr
- ggplot2
rmarkdown
========================================================
1 2 3 . . .
a <- 5
y <- f(x)
Vectors
(v <- c(1, 2, 3, 4, 5))## [1] 1 2 3 4 5(v1 <- 1:5)## [1] 1 2 3 4 5(v2 <- seq(from = 1, to = 5, by = 1))## [1] 1 2 3 4 5Vectors
( v3 <- rep(1:2, times = 3) )## [1] 1 2 1 2 1 2( v4 <- rep(1:2, each = 3) )## [1] 1 1 1 2 2 2Selecting Vector Elements
v5 <- rnorm(3)
v5## [1] 2.07436307 0.32028396 -0.09851201v5[2]## [1] 0.320284v5[-2]## [1] 2.07436307 -0.09851201Selecting Vector Elements
v5## [1] 2.07436307 0.32028396 -0.09851201v5[2:3]## [1] 0.32028396 -0.09851201v5[-(2:3)]## [1] 2.074363Selecting Vector Elements
v5## [1] 2.07436307 0.32028396 -0.09851201v5[c(1, 2)]## [1] 2.074363 0.320284v5[-c(1,2)]## [1] -0.09851201Selecting Vector Elements
v5## [1] 2.07436307 0.32028396 -0.09851201v5[v5 < 0]## [1] -0.09851201fruits <- c("apple", "pear", "apple")
fruits[fruits == "apple"]## [1] "apple" "apple"seq with Date Series
t <- seq(from = as.Date("2017-01-01"),
to = as.Date("2017-12-31"),
by = "days")
t[1:3]## [1] "2017-01-01" "2017-01-02" "2017-01-03"t2 <- seq(from = as.Date("2017-01-01"),
to = as.Date("2017-12-31"),
by = "months")
t2[1:3]## [1] "2017-01-01" "2017-02-01" "2017-03-01"Data Frame
df <- data.frame(
Fruit = c("apple", "pear", "melon"),
kg = c(100, 250, 560),
PrUn = c(2.0, 4.5, 1.2)
)
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2Read and write external data
Read
- df <- read.table(“file.txt”)
- df <- read.csv(“file.csv”)
- df <- read.csv(file.choose())
- df <- read.excel(file.choose(), sheet = “Sheet1”)
Write
- write.table(df, “file.txt”)
- write.csv(df, “file.csv”)
Selecting Data Frame Elements
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2df[1:2, ]## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5df[1, 1]## [1] apple
## Levels: apple melon pearSelecting Data Frame Elements
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2df[ , 'Fruit']## [1] apple pear melon
## Levels: apple melon pearAdd column in Data Frame
df['tot'] <- df['kg'] * df['PrUn']
df## Fruit kg PrUn tot
## 1 apple 100 2.0 200
## 2 pear 250 4.5 1125
## 3 melon 560 1.2 672df$tot <- df$kg * df$PrUn
df## Fruit kg PrUn tot
## 1 apple 100 2.0 200
## 2 pear 250 4.5 1125
## 3 melon 560 1.2 672Delete column in Data Frame
df['tot'] <- NULL
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2df$tot <- NULL
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2Structure of Data Frame
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2dim(df)## [1] 3 3names(df)## [1] "Fruit" "kg" "PrUn"Structure of Data Frame
df## Fruit kg PrUn
## 1 apple 100 2.0
## 2 pear 250 4.5
## 3 melon 560 1.2str(df)## 'data.frame': 3 obs. of 3 variables:
## $ Fruit: Factor w/ 3 levels "apple","melon",..: 1 3 2
## $ kg : num 100 250 560
## $ PrUn : num 2 4.5 1.2head() and tail()
df1 <- data.frame(a = sample(x = 1:100), b = rnorm(100))head(df1, 5)## a b
## 1 69 -0.81252232
## 2 98 1.02755742
## 3 29 -0.27553963
## 4 99 -0.15941451
## 5 12 0.09592824tail(df1, 5)## a b
## 96 15 -0.6747514
## 97 27 -0.5045930
## 98 63 0.2304039
## 99 38 2.8594134
## 100 90 -0.1242792plot()
head(df1)## a b
## 1 69 -0.81252232
## 2 98 1.02755742
## 3 29 -0.27553963
## 4 99 -0.15941451
## 5 12 0.09592824
## 6 60 -0.66987630plot(df1)
plot()
(df1.10 <- head(df1, 10))## a b
## 1 69 -0.81252232
## 2 98 1.02755742
## 3 29 -0.27553963
## 4 99 -0.15941451
## 5 12 0.09592824
## 6 60 -0.66987630
## 7 76 1.08198439
## 8 84 1.56556105
## 9 87 -0.41985035
## 10 40 1.18134961plot(df1.10$a, type = "l", col="red")
summary()
summary(df1)## a b
## Min. : 1.00 Min. :-2.23800
## 1st Qu.: 25.75 1st Qu.:-0.61497
## Median : 50.50 Median :-0.14508
## Mean : 50.50 Mean : 0.01599
## 3rd Qu.: 75.25 3rd Qu.: 0.79671
## Max. :100.00 Max. : 3.26078Linear Regression()
lr <- lm(formula = a ~ b, data = df1)
lr##
## Call:
## lm(formula = a ~ b, data = df1)
##
## Coefficients:
## (Intercept) b
## 50.525 -1.548Linear Regression()
par(mfrow=c(1,2))
plot(lr)
Mean Square with %>%
v <- c(22, 3, 2, 5, 6, 77, 8, 11, 9, 7, 57)sqrt(mean(v^2))## [1] 30.22792library(magrittr)
v^2 %>%
mean() %>%
sqrt()## [1] 30.22792dplyr
select() select columns
filter() filter rows
summarise() summarise values
arrange() re-order or arrange rows
mutate() create new columns
group_by() group data
dplyr
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, uniondf_a <- data.frame(year = rep(2013:2017, each = 12),
month = rep(1:12, times = 5),
num = sample(1:20),
weight = runif(n = 60, min = 1.0, max = 10.0))
head(df_a, 4)## year month num weight
## 1 2013 1 5 5.145590
## 2 2013 2 3 4.459135
## 3 2013 3 7 5.257585
## 4 2013 4 16 6.867127dplyr - select()
df_a_select <- select(df_a, year, num, weight)
head(df_a_select)## year num weight
## 1 2013 5 5.145590
## 2 2013 3 4.459135
## 3 2013 7 5.257585
## 4 2013 16 6.867127
## 5 2013 2 8.802364
## 6 2013 20 2.425743dplyr - filter()
df_a_filter <- filter(df_a, month == 2, num > 3)
head(df_a_filter)## year month num weight
## 1 2014 2 10 6.463854
## 2 2015 2 20 5.334409
## 3 2016 2 19 3.020158dplyr - summarise()
df_a_summarise <- summarise(df_a,
avg_n = mean(num),
avg_w = mean(weight))
head(df_a_summarise)## avg_n avg_w
## 1 10.5 5.438779dplyr - arrange()
df_a_arrange <- arrange(df_a, desc(num))
head(df_a_arrange)## year month num weight
## 1 2013 6 20 2.425743
## 2 2015 2 20 5.334409
## 3 2016 10 20 7.893768
## 4 2014 6 19 4.861062
## 5 2016 2 19 3.020158
## 6 2017 10 19 7.413796dplyr - mutate()
df_a_mutate <- mutate(df_a, tot = num * weight)
head(df_a_mutate)## year month num weight tot
## 1 2013 1 5 5.145590 25.72795
## 2 2013 2 3 4.459135 13.37741
## 3 2013 3 7 5.257585 36.80310
## 4 2013 4 16 6.867127 109.87404
## 5 2013 5 2 8.802364 17.60473
## 6 2013 6 20 2.425743 48.51485dplyr - group_by() - summarise() with %>%
df_a_gr <- df_a %>%
group_by(year) %>%
summarise(avg_n = mean(num),
avg_w = mean(weight))
head(df_a_gr)## # A tibble: 5 x 3
## year avg_n avg_w
## <int> <dbl> <dbl>
## 1 2013 8.666667 5.271529
## 2 2014 11.416667 5.853483
## 3 2015 10.583333 5.285630
## 4 2016 9.750000 6.315981
## 5 2017 12.083333 4.467274dplyr - group_by() - summarise_all() with %>%
df_a_gr <- df_a %>%
group_by(year) %>%
summarise_all(funs(mean))
head(df_a_gr)## # A tibble: 5 x 4
## year month num weight
## <int> <dbl> <dbl> <dbl>
## 1 2013 6.5 8.666667 5.271529
## 2 2014 6.5 11.416667 5.853483
## 3 2015 6.5 10.583333 5.285630
## 4 2016 6.5 9.750000 6.315981
## 5 2017 6.5 12.083333 4.467274ggplot2
library(ggplot2)
t <- seq(from = as.Date("2017-01-01"),
to = as.Date("2017-12-31"),
by = "days")
val <- rnorm(365)
df <- data.frame(tempo = t, valori = val)
f <- ggplot(df,aes(x = tempo, y = valori)) +
geom_line() +
theme_bw()f 
ggplot2 - examples
start=as.Date("2017-01-01")
df9 <- data.frame(date = seq(from = start, to = start+99, by="days"),
descr = sample(x = c("cat","dog","mouse"), size = 100, replace = T),
val = sample(x = 1:30, size = 100, replace = T))
head(df9, 40)## date descr val
## 1 2017-01-01 mouse 3
## 2 2017-01-02 cat 13
## 3 2017-01-03 cat 11
## 4 2017-01-04 dog 21
## 5 2017-01-05 mouse 21
## 6 2017-01-06 cat 3
## 7 2017-01-07 cat 1
## 8 2017-01-08 cat 30
## 9 2017-01-09 dog 27
## 10 2017-01-10 cat 22
## 11 2017-01-11 mouse 13
## 12 2017-01-12 cat 18
## 13 2017-01-13 mouse 18
## 14 2017-01-14 cat 14
## 15 2017-01-15 cat 23
## 16 2017-01-16 dog 30
## 17 2017-01-17 cat 16
## 18 2017-01-18 dog 26
## 19 2017-01-19 dog 22
## 20 2017-01-20 mouse 21
## 21 2017-01-21 mouse 22
## 22 2017-01-22 cat 1
## 23 2017-01-23 dog 20
## 24 2017-01-24 cat 25
## 25 2017-01-25 mouse 7
## 26 2017-01-26 mouse 28
## 27 2017-01-27 dog 10
## 28 2017-01-28 dog 19
## 29 2017-01-29 dog 1
## 30 2017-01-30 mouse 13
## 31 2017-01-31 mouse 19
## 32 2017-02-01 cat 4
## 33 2017-02-02 mouse 5
## 34 2017-02-03 dog 5
## 35 2017-02-04 cat 17
## 36 2017-02-05 mouse 1
## 37 2017-02-06 cat 26
## 38 2017-02-07 cat 30
## 39 2017-02-08 mouse 25
## 40 2017-02-09 mouse 22ggplot2 - examples
g <- ggplot(df9, aes(descr, val))
gg <- ggplot(df9, aes(x=val))
gg2 <- ggplot(df9, aes(x=val, fill=descr))
gg3 <- ggplot(df9, aes(x=descr, y=val, fill=descr))g + geom_violin()
g + geom_violin()
g + geom_violin(aes(fill=descr))
g + geom_violin(aes(fill=descr))
g + geom_violin() + geom_jitter()
g + geom_violin() + geom_jitter(height=0,width=0.1)
g + geom_violin(draw_quantiles = c(0.25, 0.5, 0.75))
g + geom_violin(draw_quantiles = c(0.25, 0.5, 0.75)) 
gg + geom_histogram()
gg + geom_histogram() ## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
gg2 + geom_histogram()
gg2 + geom_histogram()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
gg2 + geom_density(alpha = .3)
gg2 + geom_density(alpha = .3)
gg2 + geom_density(alpha=.3) + facet_grid(descr ~.)
gg2 + geom_density(alpha=.3) + facet_grid(descr ~.)
gg3 + geom_boxplot(alpha = .3)
gg3 + geom_boxplot(alpha = .3)
gg3 + geom_boxplot() + stat_summary()
gg3 + geom_boxplot(alpha = .3) +
stat_summary(fun.y = mean, geom = "point", shape = 5, size = 4)
========================================================
“Analizzare i dati con R” Linux Day a Pisa 2017
Author: Alessandra Santi
E-mail: santi.info@gmail.com
License: CC-BY-SA