I am in the process of building a simple model of the UK Retail Market as a Predator-Prey System. The literature review phase is now underway.
Initially the situation will be modelled the from several viewpoints and then triangulate them. The approaches which I am using are:
- Modelling theoretically using the generalised Lotka–Volterra competition equations with finite carrying capacity and multiple competitors. I am developing Jupyter and Mathematica notebooks to perform this non-linear ODE numerical analysis.
- Using Agent Based Modelling (ABM) methods to model the UK retail market as a predator-Prey system. This will be performed using the Netlogo ABM environment. This method is a good check on the models arising from the Lotka–Volterra equations.
- A supervised Machine Learning forecasting model built using regression, techniques. I have written preliminary Jupyter notebooks to implement these techniques.
There appear to be a reasonable number of relevant datasets available on which to base the model. The sources of high quality UK retail data identified so far are the ONS, CDRC and the BRC. If there are any other high quality datasets available I would be delighted to be informed of them and allowed access.
I am actively seeking support with regard to this project particularly from Retail organisations themselves, so if you feel this project might be of some interest or value to your organisation please get in contact. Ongoing updates about the project will appear here and on Academia. The software developed will be deposited at Github.