This jupyter notebook is essentially a crash course in Bayesian statistics where I try to cover the most important probabilistic ideas and numerical methods for the Master/PhD course Geophysical Data Science GEO4300/GEO9300 led by Norbert Pirk @norberp at the University of Oslo. It corresponds to a 2 hour guest lecture that I give in this course, but the notebook should be completely self-contained and is thus arguably well suited for self-study. The geophysical data science course we offer uio.no/studier/emner/matnat/geofag/GEO4300/ covers wide range of data science topics with relevance to research in geophysics of the entire Earth system (not just fossils) and beyond while being almost entirely based on modern and open programming tools in the form of jupyter notebooks and the Python programming language.
Please reach out to me Kristoffer Aalstad ([email protected]) if you spot any typos, errors, or have any feedback.
Freely available as pdfs online:
Evensen et al. (2022): Data Assimialtion Fundamentals, doi:10.1007/978-3-030-96709-3
Gelman et al. (2013): Bayesian Data Analysis (3rd ed), doi:10.1201/b16018
Murphy (2023): Probabilistic Machine Learning: Advanced Topics, probml.github.io/pml-book/book2.html
Särkkä and Svensson (2023): Bayesian Filtering and Smoothing doi:10.1017/9781108917407
As far as I know not free, but also great resources:
Otsuka (2023): Thinking about Statistics, doi:10.4324/9781003319061
McElreath (2020): Statistical Rethinking (2nd ed), doi:10.1201/9780429029608
I would also warmly recommend checking Richard McElreath's fantastic lectures on YouTube.
We kindly thank all open data providers. Data from the Blindern (i.e. at the Uni. Oslo campus) meteorological station (SN18700) was obtained from Norwegian Meteorological Institute through https://seklima.met.no/. GISS Surface Temperature Analysis (GISTEMP) data was obtained from the NASA Goddard Institute for Space Studies via https://data.giss.nasa.gov/gistemp/.