Introduction to ggplot2 Package

All the graphics generated using ggplot2 package follow common design philosophy – A Grammar of Graphics.

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. Facetting can be used to generate the same plot for different subsets of the dataset. It is the combination of these independent components that make up a graphic.

The Rmarkdown file used in the session can be found at:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/commit/d92ba40d9717b60046a185a63e5f973d57271daf

0 0

Scatter plots are used to graphically represent the relationship between the two variables. The values of the variables are represented by dots. The positioning of the dots on the vertical and horizontal axis will inform the value of the respective data point; hence, scatter plots make use of Cartesian coordinates to display the values of the variables in a data set.

In case of scatter plot, one may encounter an issue of over-plotting – that is several points are plotted on top of each other over and over again. There are two ways to deal with over-plotting issue:
– Changing the transparency of points
– Adding small random noise to each data point
In this video, we will have a look at both these methods.

Finally, one can overlay a trend line on scatter plot using geom_smooth ( ) layer.

We will work on IMDB movies data set and check whether a relationship exists between movie duration and gross revenue been generated by the movie.

The Rmarkdown file used in this video can be found at:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/blob/main/scatter_plot.Rmd

0 0

A bar graph is a frequently used type of display that compares counts, frequencies, totals or other summary measures for a series of categories, e.g. sales in different market sectors or in quarters in a financial year. The bar graph can be laid out with the categories either on the vertical or horizontal axis of the display – depending on whether we consider making a vertical or horizontal comparison is easier for interpreting the graph.

In ggplot2, the bar graph is created using geom_bar ( ) object. ggplot ( ) function uses statistical transformations (stat="count") of the data frame to create frequency bar graph i.e. counting number of occurrences the categorical variable in a data set.
The statistical transformation (stat="identity") is used to plot the raw values of the categorical variable.

We will work on IMDB movies data set and create bar chart to understand more on the movies' Genre.
The Rmarkdown file is available at Github repo:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/blob/main/bar%20plot.Rmd

0 0

Stacked bar charts usually represent a series of bars or columns stacked on top of one another. They are widely used to effectively portray comparisons of total values across several categories.

For e.g. We can show the year on year product sale using simple bar chart, but we can also add geography variable to it and show the same information using stacked bar chart..with each stack in the bar showing one particular geography….this way we can check the sale trend across various geographies.

Proportionate stacked bar chart shows relative proportions for each category standardized to 100% scale.

We will use Titanic Data set and try to come up with the answers to following questions using stacked bar and proportionate stacked bar charts.

– Out of total male and female passengers on the Titanic ship, how many of them are survived?
– Does survival rate of passengers has any bearing on its socio-economic class?
– What is survival percentage of male and female passengers?
– What is survival percentage of passengers based on socio-economic class?

We will also create contingency tables and use those in creating stacked bar charts.

The link for the first part video – How to draw bar charts is below:
https://youtu.be/wUWwL5ksIXM

The Github Repo for Rmarkedown file and Titanic Data set is:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/tree/main

0 4

Histograms are graphs of a distribution of single continuous variable designed to show centering, dispersion (spread), and shape (relative frequency) of the data. Histograms can provide a visual display of large amounts of data that are difficult to understand in a tabular, or spreadsheet form.

In this video, we will work on two datasets – rivers data – which is about length in miles of 141 major rivers in North America, and geyser data – which is eruptions data from the ‘Old Faithful’ geyser in Yellowstone National Park.

We will use geom_histogram ( ) geometric object to plot the histogram. We will also learn how to change the layout of histogram using additional arguments to geom_histogram ( ) function – such as boundary and binwidth argument.

Finally, we will use frequency polygon to show the distribution of the data by plotting a smooth curve on histogram.

The Rmakrdown file is used in this video is available at Github repo:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R

0 0

lubridate package API:
https://lubridate.tidyverse.org/reference/ymd.html

Rmarkdown and Data-set files used in this video:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R

0 0

A heat map (or heatmap) is a 2-dimensional data visualization technique that represents the magnitude of individual values within a dataset as a color. By observing how cell colors change across each axis, you can observe if there are any patterns in value for one or both variables.
In this video, we will work upon a Flights data-set from Nycflights13 package. We will show the flight delays for Delta Airlines carriers across all its routes using Heat map.

Correlogram displays the correlation coefficient for all possible combinations of two variables in the data set. We will combine Flights and Weather data-set from NYCflights13 package, and create correlogram to check, whether any relationship exists between flight delays and various weather parameters.

The R-markdown file is uploaded on Github repository:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/blob/main/heat_map_youtube.Rmd

0 0

Faceting is data visualization technique, that partitions the plot into a matrix of panels. Each panel shows a different subset of the data. The subset of data is generated by dividing the main data set by some discrete or categorical variable.

There are two types of faceting:
– facet_wrap( )
– facet_grid( )

– facet_grid( ) produces 2D grid of panels, and uses all combinations between two categorical variables to create a sub-plots. Thus, if the data for some combinations is not available, it leads to empty sub-plot.

– facet_wrap( ) produces 1D chart but wraps it in 2D. Since it is essentially 1D chart, there are no empty sub-plots created using facet_wrap( ) function.

we will use gapminder data-set to create facets. We will also use some useful techniques that can be used to modify the facets in terms of Axis scales and plot layout.

the Rmarkdwon file used in this video can be found at Git repo:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/blob/main/facet_diagrma_version1.Rmd

3 0

When constructing a data visualization, it is often necessary to make annotations to the data displayed. Annotations are nothing but the meta-data for the plot. They provide additional contextual information to the plot such as:
– Char title
– Axis title
– Text labels
– Legends
– Axis tick mark labels and their placements
– Colors in plots
In this video, we will use gapminder data and annotate its non-data components using various functions from ggplot2 package used for the annotation.

– geom_text ( ) function is used to annotate the chart
– annotate ( ) function is used to annotate only the outlier in the chart
– labs ( ) function is used for annotations of axis title, legends and chart title.
– scale_ functions are used to change the axis scales
– RcolorBrewer package is used to change the colors in the plot

The Rmarkdown file used in this video can be found at Github Repository:
https://github.com/siddharth-sahasrabudhe/ggplot2-package—data-visualisation-in-R/blob/main/axis_legend_version1.Rmd

0 1