BayesianEcosystems_IAP

Notes and code for Bayesian ecosystem modeling IAP course

Schedule

Session #1: Jan 16

• Installing PyStan
• Introduce ecosystem equations
• Simulate from the ecosystem model
• Generate synthetic observations
• First fit of the model to data in Stan

Session #2: Jan 23

• How to plot and analyze output from a Bayesian analysis
• How to make predictions from a Bayesian model
• Analyze how parameter and prediction uncertainty varies with size and dimension of observational dataset

Session #3: Jan 30

• Extend the plankton ecosystem model to include a fish population
• Fit the model to plankton + fish populations
• Use the fitted model to make predictions (and quantify uncertainty!) about the amount of fishing the ecosystem can support

Installing PyStan

If you are using Anaconda, it recommended that you first update conda by issuing

conda update --all

and then issue

conda install pystan

If you are not using Anaconda, issue

pip install pystan

Installing Jupyter Notebooks

Jupyter notebooks should already be installed if you are using Anaconda. You should be able navigate to the directory containing your notebooks and issue

jupyter notebook

to launch the interactive notebooks in a browser window.

If you are using Python 3 without Anaconda, you can install jupyter notebooks via

pip3 install jupyter

If you are using Python 2 without Anaconda, you can install jupyter notebooks via

pip install jupyter

then you can navigate to the directory containing your notebooks and issue

jupyter notebook