search

janboone / applied-economics

course material for the course applied economics
17 stars 21 forks source link

Can't do the Model questions #37

Closed florisvanlaarhoven closed 3 years ago

florisvanlaarhoven commented 3 years ago

Hi,

I was wondering how important it is to do the last section of healthcare?

with model_2014: trace_2014 = pm.sample(n_samples,step = pm.Metropolis(),start = pm.find_MAP())

--> As of this code, I can't work on the assignment anymore, because running this cell takes too long (only 0.2% was loaded after 5 minutes, even when I brought down n_samples to 1).

lindapelzer commented 3 years ago

See the announcement on Canvas: IT cannot solve this in time. Hence you can stop before the section "reversing the probability distribution".

janboone commented 3 years ago

That is indeed weird; I will ask IT what is happening.

The last section is not important to do. The main point is to get the data on the jupyter server, read it into a dataframe and then get it into the right format.

The Bayesian analysis is just to show you what is possible in python; but then, of course, it should actually work.