ericmjl/bayesian-stats-modelling-tutorial

Distributions as 1st class citizens?

Closed this issue · 3 comments

ericmjl commented

Could there be a restructuring of NB2's pre-PPL content, such that the "distributions as 1st-class citizens" could be emphasized? I have a hunch that this might help with learning.

I suspect that maybe using scipy.stats could help? It helps reinforce the idea that:

  1. PDFs are nothing more than just math functions.
  2. Calculation of likelihood is all "just" functions.
  3. Math can be abstracted.
hugobowne commented

lets discuss this. wrt your point 1, a major reason for doing this by simulation is to take an alternative path to the math, with the aim of lowering the barrier to entry.

also remember the carpentries' & @gvwilson's thoughts on using the term 'just' in pedagogy! I recall you'd be fined a dollar for each use of it! :D

i definitely believe we could emphasize "distributions as 1st-class citizens" as much as we want. it's such an important take-away. i'm not sure scipy.stats is the answer.

i think simulation and probabilistic storytelling is viable path here.

hugobowne commented

@ericmjl what do you mean by "Math can be abstracted"?

ericmjl commented

This one was also post-PyCon, so I think it’s old.