Open Science Synthesis for the Delta Science Program: Week 2
0.1
Schedule
0.1.1
Code of Conduct
0.1.2
Logistics
0.1.3
About this book
1
Session 1: Re-Introductions and Setup
1.1
Datasets of Interest
2
Session 2: git conflicts
2.1
Learning Objectives
2.2
Introduction
2.3
Collaborating with a trusted colleague
without conflicts
Setup
2.3.1
Step 1: Collaborator clone
2.3.2
Step 2: Collaborator Edits
2.3.3
Step 3: Collaborator commit and push
2.3.4
Step 4: Owner pull
2.3.5
Step 5: Owner edits, commit, and push
2.3.6
Step 6: Collaborator pull
Challenge
2.4
Merge conflicts
2.5
How to resolve a conflict
Abort, abort, abort…
Checkout
Pull and edit the file
2.5.1
Producing and resolving merge conflicts
Merge Conflict Challenge
2.6
Workflows to avoid merge conflicts
2.7
Collaborating using Git
2.7.1
Learning Objectives
2.7.2
Pull requests
2.7.3
Exercise: Create and merge pull requests
2.7.4
Branches
3
Session 3: Collaborative Synthesis
4
Session 4: Introduction to Bayesian modeling
4.1
Learning Objectives
4.2
Why choose Bayesian?
4.2.1
Philosophically sound and consistent
4.2.2
Flexible
4.2.3
Clear inference
4.2.4
Uses all available information
4.3
Review of probability, Bayes’ rule, and distributions
4.3.1
Probabability
4.3.2
Bayes’ rule
4.3.3
Bayesian inference
4.3.4
Quick review of distributions
4.3.5
Selecting priors
4.4
Drafting Bayesian models
4.4.1
Graphical representations of hierarchical models
4.5
Simple hierachical example: snow fences
4.5.1
Acknowledgements
5
Session 5: Collaborative Synthesis
6
Session 6: Implementing Bayesian models
6.1
Learning Objectives
6.2
Overview of the Bayesian modeling process
6.3
Markov chain Monte Carlo
6.3.1
Inference from iterative simulation
6.4
Programing statistical models
6.4.1
Compiling a JAGS model
6.4.2
Sampling the posterior distribution
6.5
Evaluating model diagnostics
Interactive problem: Logistic regression model
6.5.1
Acknowledgements
7
Session 7: Collaborative Synthesis
8
Session 8: Collaborative Synthesis
9
Session 9: Informative Priors
9.1
Priors
9.1.1
Learning Objectives
9.1.2
Background
9.1.3
Conjugate Priors
9.1.4
Uniformative / Weak / Flat Priors
9.1.5
Regularizing priors
9.1.6
Informative Priors
9.1.7
Generating Priors based or Expert knowledge:
9.1.8
Statistical fitting to data
9.1.9
References and further reading
10
Session 10: Missing Data
10.1
Missing Data
10.1.1
General Concepts
10.1.2
Where does missing data come from (or go?)
10.1.3
What to do about missing data
10.1.4
Types of missing data
10.1.5
Approaches
10.1.6
References
11
Session 11: Time series and forecasting
Published with bookdown
Open Science Synthesis for the Delta Science Program: Week 2
11
Session 11: Time series and forecasting