The idea of the project is to corroborate certain macroeconomic principles such as the relationship between the idea of relative wealth and spending behaviour.
To do so, we to study the evolution of inflation versus household consumption and house price index and household consumption. The aim would then be to calculate the latency in the changes in the dependent trend (household consumption) in response to those in the inflation rate / house price index.
This would effectively help us understand the normal population's actual behaviour in relation to what macro-economic principles predict.
We then propose to check this for countries and see if there is a correlation between the average economic level for a given country and consumer behaviour.
We propose to do this for 10 countries including Switzerland, Italy, Germany, India and Kenya (whose data is relatively complete for the chosen time period with the world bank c.f. http://databank.worldbank.org/data/home.aspx)
In order to compare these changes and visualise them in a comprehensive manner, while showing SIX dimensions was the biggest challenge of this project.
The six dimensions are:
- Year : Hover on data bubble
- Percentage Change in X axis variable: X coordinate
- Percentage Change in Y axis variable: Y coordinate
- Absolute value of X axis variable: Proportional to radius (Hover on data bubble)
- Absolute value of Y axis variable: Proportional to radius (Hover on data bubble)
- Countries : Choose from list
Watch the Demo(gif) above and follow the steps to play with the server yourself! (commands for Mac/Linux)
(Or)
Follow the steps below:
To run the visualisation download the Bokeh_server folder onto your machine.
Change directory into the Bokeh_server and run bokeh command (bokeh serve --show main.py).(http://bokeh.pydata.org/en/latest/docs/user_guide/server.html)
- Choose the X and Y axis variables as desired and hit play to see time evolution.
- Hit reset button on the graph if you change variables while running the animation.
- The server looks best when visualised full screen and with inverted colours!