02.Causation01.Causal Bayesian networks - sporedata/researchdesigneR GitHub Wiki

1. Use cases: in which situations should I use this method?

2. Input: what kind of data does the method require?

  1. Pre-requisits for causal-analysis

3. Algorithm: how does the method work?

Model mechanics

Describing in words

While static Bayesian Networks deal with cross-sectional data, Dynamic Bayesian Networks (DBN) model time series data with an auto-regression component - see chapter 3 of [2]. Of importance, DBN are first-order Markovian, meaning that the graph only remembers the previous node but not others. DBN also allows for feedback loops, which are often present in social, psychological, and biological phenomena.

Assumptions of DBN

    1. The time component of DBN allow for feedback loops, which are not possible under traditional Bayesian networks (intervals are constant)
    1. The process is stochastic markovian, or the events only depend on the previous event.
    1. Any variable at time t is dependent on the past variables only through the variables observed at time (t − 1).
    1. The rows in the dataset are not transformations of other rows [5]

Describing in images

Describing with code

Breaking down equations

Data science packages

  • bnlearn for R [3]

Suggested companion methods

Learning materials

  1. Books
    • Bayesian Networks: With Examples in R [4].
    • Bayesian Networks in R: with Applications in Systems Biology [5].
  2. Articles

4. Output: how do I interpret this method's results?

Typical tables and plots and corresponding text description

Metaphors

Networks displaying the connections between causes and effects.

Reporting guidelines

Bayesian methods in health technology

5. SporeData-specific

Templates

Data science functions

References

[1] Zandonà A, Vasta R, Chiò A, Di Camillo B. A Dynamic Bayesian Network model for the simulation of Amyotrophic Lateral Sclerosis progression. BMC bioinformatics. 2019 Apr 1;20(4):118.

[2] Nagarajan R, Scutari M, Lèbre S. Bayesian Networks in R with Applications in Systems Biology 2013. New York, NYSpringer-Verlag.

[3] Scutari M. bnlearn-an R package for Bayesian network learning and inference. UCL Genetics Institute, University College, London, London, UK. 2011.

[4] Scutari M, Denis JB. Bayesian Networks: With Examples in R. CRC press; 2014 Jun 20.

[5] Nagarajan R, Scutari M, Lèbre S. Bayesian Networks in R: with Applications in Systems Biology. Springer. 2013;122:125-7.

[6] Razavi M, Glasziou P, Klocksieben FA, Ioannidis JP, Chalmers I, Djulbegovic B. US Food and Drug Administration Approvals of drugs and devices based on nonrandomized clinical trials: a systematic review and meta-analysis. JAMA network open. 2019 Sep 4;2(9):e1911111-.

⚠️ **GitHub.com Fallback** ⚠️