## ----IntroEDA-ImportanceViz_readdata-------------------------------------
library(datasets)
str(anscombe)
# Let's reorg them into four data frames:
anscombe1 <- data.frame(anscombe$x1,anscombe$y1)
names(anscombe1) <- c("x","y")
anscombe2 <- data.frame(anscombe$x2,anscombe$y2)
names(anscombe2) <- c("x","y")
anscombe3 <- data.frame(anscombe$x3,anscombe$y3)
names(anscombe3) <- c("x","y")
anscombe4 <- data.frame(anscombe$x4,anscombe$y4)
names(anscombe4) <- c("x","y")


## ----IntroEDA-ImportanceViz_anmean---------------------------------------
paste(mean(anscombe1$x),mean(anscombe1$y))
paste(mean(anscombe2$x),mean(anscombe2$y))
paste(mean(anscombe3$x),mean(anscombe3$y))
paste(mean(anscombe4$x),mean(anscombe4$y))


## ----IntroEDA-ImportanceViz_anstddev-------------------------------------
paste(sd(anscombe1$x),sd(anscombe1$y))
paste(sd(anscombe2$x),sd(anscombe2$y))
paste(sd(anscombe3$x),sd(anscombe3$y))
paste(sd(anscombe4$x),sd(anscombe4$y))


## ----IntroEDA-ImportanceViz_ansplot, fig.width=8, fig.height=8-----------
library(ggplot2)
# Let's add a label to each dataset.
anscombe1$i <- "Set 1"
anscombe2$i <- "Set 2"
anscombe3$i <- "Set 3"
anscombe4$i <- "Set 4"
# Now we can merge the datasets.
# https://stackoverflow.com/questions/16138693/rbind-multiple-data-sets
allData <- rbind(anscombe1,anscombe2,anscombe3,anscombe4)
# Create a plot with a linear regression plot over the points.
# https://stackoverflow.com/questions/15633714/adding-a-regression-line-on-a-ggplot
p <- ggplot(data = allData, aes(x = x, y = y)) + geom_point() + 
     geom_smooth(method='lm',fullrange=TRUE)
# Plot wrapped using the i column.
# https://www3.nd.edu/~steve/computing_with_data/13_Facets/facets.html
p + facet_wrap(~i)