data.frame
“Grammatik för grafik”
data
: en data.frame med all dataaes
: aesthteic mappings (färger, former)geom
: geometriska objektfacets
: skapar subplottar baserat på villkorscales
: skalor för aesthteic mappingscoordinate system
Bild från denna källa
print()
ggplot()
+
”,
+ geom_line()
”In brief, the grammar tells us that a statistical graphic is a mapping from data to aesthetic attributes (colour, shape, size) of geometric objects (points, lines, bars). The plot may also contain statistical transformations of the data and is drawn on a specific coordinate system.”
Från “ggplot2 book” av Hadley Wickham
aes |
Beskrivning |
---|---|
x |
x-axel |
y |
y-axel |
size |
storlek |
color |
färg |
shape |
form |
geom |
Beskrivning |
---|---|
geom_point |
Scatterplot |
geom_line |
Line graph |
geom_bar |
Barplot |
geom_boxplot |
Boxplot |
geom_histogram |
Histogram |
geom |
aes |
---|---|
geom_points |
point shape, point size |
geom_line |
line type, line size |
geom_bar |
y min, y max, fill color, outline color |
ggplot(data=Nile) + aes(x=years, y=level) + geom_point()
ggplot(data=Nile) + aes(x=years, y=level) + geom_line()
p <- ggplot(data=Nile) + aes(x=years, y=level, color=period) + geom_point(aes(shape=period))
p
p <- ggplot(data=Nile) + aes(x=years, y=level, color=period) + geom_line(aes(type=period)) + geom_point()
p
p <- ggplot(data=Nile) + aes(x=years, y=level) + facet_grid(period~.) + geom_line()
p
p <- ggplot(data=Nile) + aes(x=years, y=level) + facet_grid(~period) + geom_line()
p
p + theme_bw()
p + theme_classic()
p + theme_grey()
qplot()
: Liknar plot()
från basgrafiken.ggplot()
chisq.test()
, fisher.test()
t.test()
cor()
och cov()
och testas medcor.test()
data("chickwts")
horsebean <- chickwts$weight[chickwts$feed == "horsebean"]
sunflower <- chickwts$weight[chickwts$feed == "sunflower"]
mean(horsebean)
[1] 160.2
mean(sunflower)
[1] 328.9167
t.test(horsebean,alternative="two.sided", mu = 150, conf.level = 0.95)
One Sample t-test
data: horsebean
t = 0.83507, df = 9, p-value = 0.4253
alternative hypothesis: true mean is not equal to 150
95 percent confidence interval:
132.5687 187.8313
sample estimates:
mean of x
160.2
t.test(horsebean, sunflower,alternative="two.sided", mu = 0, conf.level = 0.95)
Welch Two Sample t-test
data: horsebean and sunflower
t = -9.0449, df = 19.964, p-value = 1.69e-08
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
-207.6313 -129.8021
sample estimates:
mean of x mean of y
160.2000 328.9167
y ~ x1 + x2
Används på flera håll i R (ex. facet-grid)library(MASS)
library(car)
data(Prestige)
mod1 <- lm(prestige ~ income + women + education, data=Prestige)
mod2 <- lm(prestige ~ income + women + education - 1, data=Prestige)
mod3 <- lm(prestige ~ income:women + education, data=Prestige)
summary()
anova()
Exempel:
summary(mod1)
anova(mod1)
anova(mod3, mod1, test="Chisq")
plot(mod1)
durbinWatsonTest(mod1)
qqPlot(mod1)
resid()
predict()