search

gbosquechacon / statrethink_course_in_pymc3

Statistical Rethinking course in pymc3
143 stars 44 forks source link

About model_22 from week 6 #1

Closed rosgori closed 4 years ago

rosgori commented 4 years ago

You wrote

Results are quite different from the original in R. I can't seem to locate the issue. Parameter g should be much higher, around 0.28 and its HPD interval should be completely above zero to prove the point that we see a gender advantage when in reality there is none.

They are different because the code in R has a seed (set.seed(1913)) and that changes the number for g and d. If you rerun that model (changing G, S, D, A), the numbes will be almost the same.

gbosquechacon commented 4 years ago

Oh man thanks so much! I'll make sure to update the notebook this weekend : )

gbosquechacon commented 4 years ago

Fix added to the week 6 notebook.