How to find priors intuitively

How do you choose priors under constraints?
Sometimes, as a modeler, you:
- Know you want a Beta prior, but it's not as intuitive to parametrize as a Normal distribution
- Have domain knowledge ("I need a Gamma that has 95% probability mass between 0.1 and 3") but don't know how to translate that into the parameters' distribution
In this video, we'll show you how to use the new `pymc.find_constrained_prior` function to find those priors intuitively and efficiently -- no long manual plotting back-and-forth required ;)
- Link to Jupyter notebook code (PyMC 4.0 or higher required): github.com/pymc-labs/research...
#Priors #BayesianReasoning #pymc #AlexandreAndorra #Bayesianmodeling #Intuition #PriorKnowledge #Probability #BayesianInference #bayestheorem #ProbabilisticThinking #PriorBeliefs #DataAnalysis #statistics
- Presented by: Alexandre Andorra, / alex_andorra
Do you need bespoke Bayesian modeling services or corporate workshops to increase your team's Bayesian skills? Check out what we offer at PyMC Labs at www.pymc-labs.io and reach out to [e-mail] !

[@rajns8643]

Amazing stuff!! Thanks for this wonderful library!

[@PyMCLabs]

Glad you enjoyed it!

[@davidwalton8659]

Absolutely great

[@PyMCLabs]

Thanks David!

[@GreyHatGenX]

thanks for this