R workshop i r basic (4th time)

RRWWoorrkksshhooppII
gettoknowNYCopendataportalandstarttouseR
Vivian Zhang for NYC-open-data meetup
http://www.meetup.com/NYC-Open-Data/
R Workshop I http://www.nycopendata.com/RworkshopI/index.html#1
1 of 27 6/13/14, 1:50 PM

OOvveerrvviieeww
nyc open data portal
Rstudio
R
Github
hack time
·
·
·
·
·
2/27
2 of 27 6/13/14, 1:50 PM

AAddvvaannttaaggeeooffuussiinnggRRssttuuddiioo
Easiness·
install and load R packages
keep track of R dev version
download github repositories
debug faster
-
-
-
-
3/27
3 of 27 6/13/14, 1:50 PM

ddiiaammoonnddssssuubbsseettttiinnggeexxaammppllee
require(ggplot2)
head(diamonds)
## carat cut color clarity depth table price x y z
## 1 0.23 Ideal E SI2 61.5 55 326 3.95 3.98 2.43
## 2 0.21 Premium E SI1 59.8 61 326 3.89 3.84 2.31
## 3 0.23 Good E VS1 56.9 65 327 4.05 4.07 2.31
## 4 0.29 Premium I VS2 62.4 58 334 4.20 4.23 2.63
## 5 0.31 Good J SI2 63.3 58 335 4.34 4.35 2.75
## 6 0.24 Very Good J VVS2 62.8 57 336 3.94 3.96 2.48
4/27
4 of 27 6/13/14, 1:50 PM

head(diamonds[-1, ])
## 2 0.21 Premium E SI1 59.8 61 326 3.89 3.84 2.31
## 3 0.23 Good E VS1 56.9 65 327 4.05 4.07 2.31
## 4 0.29 Premium I VS2 62.4 58 334 4.20 4.23 2.63
## 5 0.31 Good J SI2 63.3 58 335 4.34 4.35 2.75
## 6 0.24 Very Good J VVS2 62.8 57 336 3.94 3.96 2.48
## 7 0.24 Very Good I VVS1 62.3 57 336 3.95 3.98 2.47
5/27
5 of 27 6/13/14, 1:50 PM

head(diamonds[, -1])
## cut color clarity depth table price x y z
## 1 Ideal E SI2 61.5 55 326 3.95 3.98 2.43
## 2 Premium E SI1 59.8 61 326 3.89 3.84 2.31
## 3 Good E VS1 56.9 65 327 4.05 4.07 2.31
## 4 Premium I VS2 62.4 58 334 4.20 4.23 2.63
## 5 Good J SI2 63.3 58 335 4.34 4.35 2.75
## 6 Very Good J VVS2 62.8 57 336 3.94 3.96 2.48
6/27
6 of 27 6/13/14, 1:50 PM

head(diamonds[c(1, 2), ])
## 1 0.23 Ideal E SI2 61.5 55 326 3.95 3.98 2.43
## 2 0.21 Premium E SI1 59.8 61 326 3.89 3.84 2.31
7/27
7 of 27 6/13/14, 1:50 PM

names(diamonds)
## [1] "carat" "cut" "color" "clarity" "depth" "table" "price"
## [8] "x" "y" "z"
head(diamonds[, c(T, T, F, F, F, F, T, F, F, F)])
## carat cut price
## 1 0.23 Ideal 326
## 2 0.21 Premium 326
## 3 0.23 Good 327
## 4 0.29 Premium 334
## 5 0.31 Good 335
## 6 0.24 Very Good 336
8/27
8 of 27 6/13/14, 1:50 PM

names(diamonds)
## [1] "carat" "cut" "color" "clarity" "depth" "table" "price"
## [8] "x" "y" "z"
head(diamonds$carat)
## [1] 0.23 0.21 0.23 0.29 0.31 0.24
diamonds[diamonds$price == max(diamonds$price), ]
## 27750 2.29 Premium I VS2 60.8 60 18823 8.5 8.47 5.16
9/27
9 of 27 6/13/14, 1:50 PM

rreeaaddiinnggaannddssuubbsseettttiinnggddaattaaiinnRR
blank
integer
logical
character
·
include all-
·
+: include;-: exclude-
·
include TRUEs-
·
lookup by name-
Source: Hadley Wickham
10/27
10 of 27 6/13/14, 1:50 PM

ddaattaassttrruuccttuurreeiinnRR
Source: Hadley Wickham
11/27
11 of 27 6/13/14, 1:50 PM

rreeaaddiinntthheeooppeennddaattaa
read.table()
read.csv()
·
·
rodent1year <- read.csv("C:UserszhangsGoogle DriveR codeRworkshop311_Service_Requests_from_2010_
header = TRUE, sep = ",")
dim(rodent1year)
summary(rodent1year)
table(rodent1year$Borough)
12/27
12 of 27 6/13/14, 1:50 PM

With() is generic function that evaluates expr in
a local environment constructed from data.
Using ggplot2, "aes" stands for "aesthetics",
"geom"" is used to create scatterplots
pplloottddiiaammoonnddss
with(diamonds, plot(carat, price)) ggplot(diamonds, aes(x = carat, y = price)) + geom_point()
13/27
13 of 27 6/13/14, 1:50 PM

ggplot2 generates more supplicated graph than the traditional graphics package. Let us play with
some color
ggplot(diamonds, aes(x = carat, y = price, colour = cut)) + geom_point()
14/27
14 of 27 6/13/14, 1:50 PM

In stead of ﬁtting linear relation, we try to ﬁt log linear relation
Log(price) is quite linear with log(carat),Bingo!
ggplot(diamonds, aes(x = log(carat), y = log(price), colour = cut)) + geom_point()
15/27
15 of 27 6/13/14, 1:50 PM

As letters go from D to J, the diamond becomes more and more yellow. The numbers beside
"S"(small) and "VS"(very small) describe the size of "internal imperfections" in the diamonds. "IF" is
internally ﬂawless.
ggplot(diamonds, aes(x = log(carat), y = log(price), colour = cut)) + geom_point() +
facet_grid(clarity ~ color)
16/27
16 of 27 6/13/14, 1:50 PM

Let us look back to a normal scale. The bottom left panel shows price vs carat for ultimate white and
internally ﬂawless diamonds. The upper right panel shows price vs carat for most unpure(or dirtiest)
and ﬂawed diamonds.
ggplot(diamonds, aes(x = carat, y = price, colour = cut)) + geom_point() + facet_grid(clarity ~
color)
17/27
17 of 27 6/13/14, 1:50 PM

As we would expect, for the diamonds at the same level of pureness(observed by row) , the price
per carat increases faster for white stone (bottom left) than for yellow stone(bottom right). And for the
diamond at the same level of color (observed by column), the price per carat increases faster for
pure stone(bottom left) than for dirty stone(upper left).
18/27
18 of 27 6/13/14, 1:50 PM

We facet the plot by one of these factor variables--clarity.
ggplot(diamonds, aes(x = carat, y = price, colour = cut)) + geom_point() + facet_grid(clarity ~
.)
19/27
19 of 27 6/13/14, 1:50 PM

ggooooddttiippttooggeenneerraatteepplloottss
The same type of graph is used over and over again while new individual component of ggplot2 is
introduced and interpreted. It is a very effective way to display complex relationship in large,
high-dimensional data. Remember, the key is to bring in only one change each time.
Source: http://gettinggeneticsdone.blogspot.com/2010/01/ggplot2-tutorial-scatterplots-in-series.html
20/27
20 of 27 6/13/14, 1:50 PM

Last , we ﬁt line for the orginal data and for the log transformed data.The linear relation is roughly
perfect of the log transformed data if we ignore the few points at two sides of the distribution.
ggplot(diamonds, aes(x = carat, y = price)) + geom_point() + geom_smooth()ggplot(diamonds, aes(x = log(carat), y = log(price))) + geo
21/27
21 of 27 6/13/14, 1:50 PM

aammaazziinnggNNYYTTiimmeessssaammppllee
http://timelyportfolio.github.io/rCharts_512paths/
Source: Timely Portfolio and NYTimes
22/27
22 of 27 6/13/14, 1:50 PM

wwhhyyddoowweeuusseeRR
Dirk's exmaple about elegance and efficiency of R Source: Dirk Eddelbuettel
23/27
23 of 27 6/13/14, 1:50 PM

wwhhyyddoowweeuusseeRR
Dirk's exmaple about elegance and efficiency of R Source: Dirk Eddelbuettel
24/27
24 of 27 6/13/14, 1:50 PM

hhaacckkttiimmee
download an open dataset using ﬁlter
read it in to your Rstudio
check the dimensity of the dataset
decide which columns you will use
plot it!
·
·
·
·
·
25/27
25 of 27 6/13/14, 1:50 PM

RReessoouurrcceess
R in a Nutshell - Joseph Adler
The Art of R Programming - Norman Matloff
ggplot2 - Elegant Graphics for Data Analysis - Hadley Wickham
26/27
26 of 27 6/13/14, 1:50 PM

27/27
27 of 27 6/13/14, 1:50 PM

R workshop i r basic (4th time)

Recommandé

Recommandé

Contenu connexe

En vedette

En vedette (7)

Similaire à R workshop i r basic (4th time)

Similaire à R workshop i r basic (4th time) (20)

Plus de Vivian S. Zhang

Plus de Vivian S. Zhang (20)

Dernier

Dernier (20)

R workshop i r basic (4th time)