1.1 Thinking like a Bayesian 1/4

DiagrammeR::grViz("
  digraph thinking_bayesian{
  
  # node statement
  node [shape = oval]
  a [label = 'Prior'];
  b [label = 'Data'];
  c [label = 'Posterior'];
  d [label = 'New data'];
  e [label = 'Posterior'];
  f [label = 'New data']
  g [style = invisible ]
  
  # edge statement
  a -> c b -> c
  c -> e d -> e
  f-> g [style = dashed] e-> g [style = dashed]
  }")

Figure 1.1: A Bayesian knowledge-building diagram

In Bayesian analysis, the prior represents what you know before seeing the data. The posterior then represents what you know having seen the data.

Both Bayesian and frequentist share a common goal: learn from data about the world around. Both use data to fit nodels, make predictions and evaluate hypothesis