An Application of the Poisson Process and Poisson Distribution to Model Earth Asteroid Impacts
Here’s some good news: if you’ve spent hours studying a concept by reading books and class notes on the theory and you just can’t seem to get it, there’s a better way to learn. Starting with theory is always difficult and frustrating because you can’t see the most important part of a concept: how it’s used to solve problems. In contrast, learning by doing — working through problems — is more effective because it gives you context, letting you fit the technique into your existing mental framework. Moreover, studying through applications is more enjoyable for those motivated by an intrinsic desire to solve problems.
In this article, we’ll apply the concepts of a Poisson Process and Poisson Distribution to model Earth asteroid impacts. We’ll build on the principles covered in The Poisson Process and Poisson Distribution Explained, putting into practice the ideas with real-world data. Through this project, we’ll get a sense of how to use statistical concepts to solve problems and how to compare the observed results with theoretical expected outcomes.
The full code for this article (with interactive apps) can be run on mybinder.org by clicking the image. The Jupyter Notebook is on GitHub.
Probability Mass Function of Number of Asteroid Impacts