Censored data and Weibull AFT model (#3095)

* Port of jazzmuesli PR

* Add explanation and docstrings, rename notebook

* Finished censored data notebook

* Rename aft notebook to weibull aft

* Revert "Finished censored data notebook"

This reverts commit 74656e3.

* Revert accidental change

* Remove censored_data.py

* Revert "Revert "Finished censored data notebook""

This reverts commit 4c6faab.

* Touch up censored_data.ipynb

* Add survival analysis session to examples

* Add parametric survival notebook

* Update doc session

* First pass at AFT notebook

* Second pass at weibull aft notebook

* Change all Bernoullis to Potentials

* Update doc to point to weibull aft nb

* Include some more explanation

* Touch ups

* Add line to release notes

* Correct Bernoulli to Potential

Author: eigenfoo

RELEASE-NOTES.md

@@ -6,6 +6,7 @@
 
 ### New features
 
+- Add documentation section on survival analysis and censored data models
 - Add `check_test_point` method to `pm.Model`
 - Add `Ordered` Transformation and `OrderedLogistic` distribution
 - Add `Chain` transformation

docs/source/api/bounds.rst

@@ -54,7 +54,7 @@ Caveats
 
 * Bounds cannot be given to variables that are ``observed``.  To model
   truncated data, use a :func:`~pymc3.model.Potential` in combination with a cumulative
-  probability function.  See `this example <https://github.com/pymc-devs/pymc3/blob/master/pymc3/examples/censored_data.py>`_.
+  probability function.  See `this example notebook <https://github.com/pymc-devs/pymc3/blob/master/docs/source/notebooks/weibull_aft.ipynb>`_.
 
 * The automatic transformation applied to continuous distributions results in
   an unnormalized probability distribution.  This doesn't effect inference

docs/source/examples.rst

@@ -28,7 +28,6 @@ Applied
    notebooks/stochastic_volatility.ipynb
    notebooks/probabilistic_matrix_factorization.ipynb
    notebooks/rugby_analytics.ipynb
-   notebooks/survival_analysis.ipynb
    notebooks/dawid-skene.ipynb
 
 GLM
@@ -46,6 +45,15 @@ GLM
    notebooks/GLM-negative-binomial-regression.ipynb
    notebooks/GLM-hierarchical-binominal-model.ipynb
 
+Survival Analysis
+=================
+
+.. toctree::
+   notebooks/survival_analysis.ipynb
+   notebooks/bayes_param_survival_pymc3.ipynb
+   notebooks/weibull_aft.ipynb
+   notebooks/censored_data.ipynb
+
 Gaussian Processes
 ==================
 
@@ -80,7 +88,6 @@ Variational Inference
    notebooks/empirical-approx-overview.ipynb
    notebooks/normalizing_flows_overview.ipynb
 
-
 Stochastic Gradient
 ===================