Skip to contents
Visualisations
1.0
slides
Purpose of Data Visualisations: EDAs and the Narrative
The Economist on their own crimes agains data visualisation design principles
The Mechanics of Data Visualisations: Fundamentals of DataVis and the Grammar of Graphics
The Mechanics of Data Visualisations: Fundamentals of DataVis and the Grammar of Graphics
Examples from the literature
Fine-tuning your graphs: Themes, colour,...
Base R tricks
Plotly: interactive graphics etc.
Extras
Maps and spatial variations (Homicide rates)
Geo-spatial analysis in python
Building a dashboard using shiny
Graphical Research Design
Binscatter plots
Regression Discontinuity Design (RDD) examples
Regression Discontinuity Design (RDD) in action: Mortality and the MDLA in the US
Event Study Designs
Reinforcement Learning in game environments
Introduction to Pygame
DRL intro
Gallery
coding
Complete Projects
The Titanic data: Survived or not ?
The Big Mac Index
The Big Mac Index
Female representation in NBER WP series
Minimum Wages in contiguous US counties
Components
Graph types
Coding exercise: Direct Labeling for the Big Mac Index time series
Coding exercise: Direct Labeling for the Big Mac Index time series
Coding exercise: Direct labeling in time series plots
Coding exercise: Taming Spaghetti Plots
Coding exercise: Taming Spaghetti Plots
Coding exercise: Viewing figures for Dr Who episodes
Coding exercise: House prices in Detroit (EDA)
Coding exercise: Some EDAs
Maps and spatial variations
Coding exercise: Heatwaves and excess mortality in France
Global warming (taming Spaghetti Plots in the ERA5 data)
projects
The Project
Part 1: NHGIS wrangling, county population, dashboards
Part 2: Heatwave mortality in Italy
Homework
HW1: Simplifying the bar plots (stacked barplots, histogram)
HW 1: Simplifying the bar plots (stacked barplots, histogram)
HW 2: Economist-style barplots (text)
HW 2: Economist-style barplots (text)
HW 3: EDA for predicting high site visits (densities, barplots)
HW 4: IMDB ratings of The Office
HW 5: Declutter a plot (Palmer Penguines)
HW 6: Direct labeling and decluttering (corruption index)
HW 6: Direct labeling and decluttering (corruption index)
HW 7: NHGIS data explorations (warm-up)
HW 8: NHGIS data extraction using the API
HW 9: Maps, temperature and heatwave data for France using ERA5
HW 10: Commuting flows for Aix (Shiny)
Light
Dark
Auto
Gallery
gallery.Rmd
Appearence
Colour
Types
Taming spaghetti plots
Layouts
Direct labelling: Bar plots
Direct labelling: Scatter plots
Flow diagrams
Directed graphs, NN
Geo-spatial
Geo-spatial
kepler: 3d map
Choropleth
thematic maps
ERA5 temp. data
Interactive
plotly, bokeh: OLS, MSE and its gradients
Bokeh: multiple sliders (shallow NN)
plotly: choropleth_map
Shiny: Commuting flows
Shiny: Excess covid deaths
Shiny dashboard + leaflet
3D
bokeh: contour (MSE)
plotly: 3d & contour (MSE)
Gradient Descent
Re. Design
binscatter plots
RDD
PyGame, RL
Deep RL: Gridworld
Deep RL: Lunar Lander
Deep RL: Space Invaders