Resources

There are tons of resources available online that will help you in learning statistical methods. Here are a few that have come to my attention that may be particularly useful.

Note that this will be amended / added to as the semester goes on. And if you have something you find valuable that isn’t on here, please share it!

Optional, supplemental materials¶

Textbooks and lecture notes¶

Videos¶

Online courses¶

Computing¶

This course is heavily computational. The language used is Python, and most graded work is submitted as executable Jupyter notebooks.

Python basics¶

Official Python (“docs”): https://www.python.org/doc/
Official Python beginner’s guide: https://wiki.python.org/moin/BeginnersGuide

Jupyter¶

Jupyter homepage: https://jupyter.org/

Anaconda distributions of Python¶

Anaconda/conda distributions of python are the strongly recommended way to use python.

Official website: https://www.anaconda.com/
For those on Windows and for anybody new to Python, I recommend installing the Anaconda distribution on whatever computer you’ll be using for this. This installation includes the Anaconda Navigator GUI.
For those on MacOS or Linux with more programming experience, I recommend installing the lightweight “miniconda” distribution instead and learning to work with it from the command line: https://docs.conda.io/projects/miniconda/en/latest/

Managing python environments and packages¶

Benefits of using Python environments rather than a single installation:¶

Precise control for each class/project/etc. on exactly what packages you want installed. (Suppose one project you’re working on uses some old legacy code that requires Python 2, whereas at this point everything else runs Python 3.)
Via conda/anaconda, automated installation of required non-python dependencies as well
On shared computing clusters where you don’t have “root” access, ability to install packages and run things (like Jupyter!) in your home directory rather than having to go through the sysadmin
Ability to experiment with updated versions of things without worrying about breaking your stuff. For example, latest Python version, 3.12, just came out. Some packages might not have support for it yet, but you can create a dedicated “python312” environment to try it out without worrying about breaking your other stuff
Conda environments underlie modern cloud-based data science and scientific computation...Binder, cloud service providers, all of these rely on conda environments

Learning¶

Spaced repetition¶

Storage strength vs. retrieval strength: https://fivetwelvethirteen.substack.com/p/retrieval-strength-and-storage-strength https://memorystrength.netlify.app

References¶

DelSole, T., & Tippett, M. (2022). Statistical Methods for Climate Scientists. Cambridge University Press. 10.1017/9781108659055
(2019). Elsevier. 10.1016/c2017-0-03921-6
Trauth, M. H. (2022). Python Recipes for Earth Sciences. In Springer Textbooks in Earth Sciences, Geography and Environment. Springer International Publishing. 10.1007/978-3-031-07719-7
Storch, H. von, & Zwiers, F. W. (1984). Statistical Analysis in Climate Research. Cambridge University Press. 10.1017/cbo9780511612336