A/B Testing Joint Explanation

CamDavidsonPilon / Probabilistic-Programming-and-Bayesian-Methods-for-Hackers

aka "Bayesian Methods for Hackers": An introduction to Bayesian methods + probabilistic programming with a computation/understanding-first, mathematics-second point of view. All in pure Python ;)

http://camdavidsonpilon.github.io/Probabilistic-Programming-and-Bayesian-Methods-for-Hackers/

MIT License

26.51k stars 7.84k forks source link

A/B Testing Joint Explanation #453

Open bluesky314 opened 5 years ago

bluesky314 commented 5 years ago

The double_joint_log_prob in ch2 A/B testing is not explained at all. Why do we return ( rv_prob_A.log_prob(prob_A)

rv_prob_B.log_prob(prob_B)
tf.reduce_sum(rv_obs_A.log_prob(observations_A))
tf.reduce_sum(rv_obs_B.log_prob(observations_B)) )

would'nt this corrospond to finding the posterior that observations_A AND observations_B occur and not the difference? Can some one provide a derivation?

bluesky314 commented 5 years ago

Do we find the joint of prob_A,prob_B and then marginalising over to get the individual posteriors? Why not do it separately then?

Why does 'burnin' stand for in burned_prob_Atrace = posterior_probA[burnin:] ?