Bayesian analysis of COVID19 evolution in Spain
Important: Bayesian analysis of COVID19 evolution in Spain with Bayesian methods, a natural approach to inference, using PyMC as probabilistic programming language.
The notebook gets the data directly from Datatista (https://github.com/datadista/datasets/blob/master/COVID%2019/nacional_covid19.csv).
This notebook is adapted from a Packt Publishing notebook (https://github.com/PacktPublishing/Bayesian-Analysis-with-Python).
The notebook updates the number of new infections reported. We have made a GP regression of COVID19 spread in Spain to infer the evolution of the ‘curve’. Gaussian Process regression is a non-parametric approach to regression or data fitting that assumes that observed data points 𝑦 are generated by some unknown latent function 𝑓(𝑥).
We share this notebook because we think can be improve.
Note: There isn’t much data today, so there will likely be a lot of uncertainty in the hyperparameter values.