Bayesian methods are a powerful tool for understanding uncertainty in our datasets and the strengths and weaknesses of our models. I’m writing a series of blog posts about how Bayesian methods can be used in practice in oceanography.
The second post in this series shows how to estimate correlation using a Bayesian approach. The post is an interactive Jupyter notebook that you can access in two ways.
You can run the notebook from the cloud
by going to this repository and
hitting the ‘launch binder’ button in the readme below the file list.
You then hit the launch
button on the next page and wait patiently for a minute while the interactive
notebook environment is created. You then open the ‘bayes_correlation.ipynb’
notebook and you can then use it as you would a notebook on your own machine.
The first notebook ‘bayes_moc_gaussian.ipynb’ on fitting a statistical distribution to a dataset is also available in the repository.
If you’d prefer to run the notebook on your own machine, you can clone or download the same repository using the green button on the top-right of the repository page. Cloning or downloading the repository will also download the data files that are in the data sub-directory.