In statcourses/BristolVis: Data visualisation and web applications course

Load the (movies) data set from the BristolVis R package. The data can be called and viewed using:

data(bmov, package = "BristolVis")
head(bmov)

Scatter plots (15 minutes)

Let's start with some simple scatter plots using the bmov data:

1. Plot length Vs. rating using the advanced graphics package (ggplot2):

require(ggplot2)
require(plotly)
(G = ggplot(bmov, aes(Length, Rating)) + geom_point())

1. Use the cut function to generate a categorical form of the variable Year with sensible cutpoints.

summary(bmov$Year)
bmov$cat_timing = cut(bmov$Year, breaks = c(1930, 1990, 2000, 2005))

1. Plot length Vs. rating such that points are coloured using categories of your generated timing:

(G = ggplot(bmov, aes(Length, Rating, color = cat_timing)) + geom_point())

1. The default colors of the previous plot are terrible! use your own color selections to generate a better plot.

(G = G + scale_color_manual(values = c("blue", "yellow", "red")))

1. Generate an interactive plot of the plot in (4) using the plotly package and name it Fig_scatter

Fig_scatter = ggplotly(G)

1. Try to zoom in using your mouse by box selection to explore further detailed information

2. Save your interactive plot as an html file.

htmlwidgets::saveWidget(Fig_scatter, "Fig_scatter.html")

Histograms (10 minutes)

1. Use the ggplot2 to plot a histogram of the movie years restricted to data after 1980.

(G = ggplot(bmov[bmov$Year>=1980,], aes(Year)) + geom_histogram())

1. Produce the same plot as in (1), but set the number of bins to 25.

(G = G + geom_histogram(bins = 25))

1. Generate an interactive plot of the plot in (2) using the plotly package and name it Fig_hist

Fig_hist = ggplotly(G)

1. Try to zoom in using your mouse by box selection to explore further detailed information and reset the plot (double-click)

2. Save your interactive plot as an html file.

htmlwidgets::saveWidget(Fig_hist, "Fig_hist.html")

Boxplots (10 minutes)

1. Generate a boxplot for the ratings data by generated categories of production timing using ggplot2.

(G = ggplot(bmov, aes(x=cat_timing, y =Rating)) + geom_boxplot(aes(group = cat_timing)))

1. Try generating a similar interactive boxplot

Fig_box = ggplotly(G)

1. save the interactive plot to an html file

htmlwidgets::saveWidget(Fig_box, "Fig_box.html")

Corrlation matrix (20 minutes)

1. Use the built-in iris data to compute a correlation matrix and correlation p-values for the continous (first four) variables

data(iris)
head(iris)
require(ggcorrplot)
data_cont = iris[,1:4]
Corr = cor(data_cont)
corr.p = cor_pmat(data_cont)

1. Visualize the correlation matrix using the method = "circle"

(G = ggcorrplot(Corr, method = "circle"))

1. Show the lower triangle using hierarchical clustering and square method rather than circle.

(G = ggcorrplot(Corr, hc.order = TRUE, type = "lower"))

1. Use correlation significance level 0.01 to highlight the non-significant coefficient

(G = ggcorrplot(Corr, hc.order = TRUE, type = "lower", p.mat = corr.p, sig.level = 0.01))

1. Use different shape - rather than the cross - to highlight the non-significant coefficient (use help of the function ggcorrplot by typing: ?ggcorrplot to find out)

(G = ggcorrplot(Corr, hc.order = TRUE, type = "lower", p.mat = corr.p, sig.level = 0.01, pch = 10))

1. Add coeficient values on the plot in (3)

(G1 = ggcorrplot(Corr, hc.order = TRUE, type = "lower", lab = TRUE, lab_size = 3))

7.Produce interactive plot of the plot in (6)

Fig_cor = ggplotly(G1)

7.save the interactive plot in as html.

htmlwidgets::saveWidget(Fig_cor, "Fig_cor.html")

statcourses/BristolVis documentation built on Jan. 31, 2021, 9:24 p.m.