Some adjustments in docs

OpenSourceEconomics · Aug 28, 2018 · 67d3e39 · 67d3e39
1 parent f124a2b
commit 67d3e39
Show file tree

Hide file tree

Showing 20 changed files with 1,516 additions and 21 deletions.
diff --git a/docs/conf.py b/docs/conf.py
@@ -43,7 +43,8 @@
     'sphinx.ext.mathjax',
     'sphinx.ext.ifconfig',
     'sphinx.ext.viewcode',
-    'sphinxcontrib.bibtex'
+    'sphinxcontrib.bibtex',
+    'sphinx.ext.imgconverter'
 ]
 
 # Add any paths that contain templates here, relative to this directory.

diff --git a/docs/credits.rst b/docs/credits.rst
@@ -3,7 +3,7 @@ Contact and Credits
 
 If you have any questions or comments, please do not hesitate to contact us directly.
 
-.. image:: https://badges.gitter.im/Join%20Chat.svg
+.. image:: figures/Join_Chat.svg
    :target: https://gitter.im/OpenSourceEconomics/Lobby?utm_source=share-link&utm_medium=link&utm_campaign=share-link
 
 Development Lead

diff --git a/docs/economics.rst b/docs/economics.rst
@@ -57,7 +57,7 @@ The information available to the econometrician and the agent determines the set
 Essential Heterogeneity
 ^^^^^^^^^^^^^^^^^^^^^^^^
 
-If agents select their treatment status based on benefits unobserved by the econometrician (selection on unobservables), then there is no unique effect of a treatment or a policy even after conditioning on observable characteristics. Average benefits are different from marginal benefits, and different policies select individuals at different margins. Conventional econometric methods that only account for selection on observables, like matching (:cite:`CocRub72`, :cite:`RoRu1983`, :cite:`HeIcSmTo98`), are not able to identify any parameter of interest (:cite:`HecVyr05`, :cite:`HeUrVy06`). For example, Carneiro (2011) present evidence on agents selecting their level of education based on their unobservable gains and demonstrate the importance of adjusting the estimation strategy to allow for this fact. Heckman and Schmierer (:cite:`Heckman2010`) propose a variety of tests for the the presence of essential heterogeneity.
+If agents select their treatment status based on benefits unobserved by the econometrician (selection on unobservables), then there is no unique effect of a treatment or a policy even after conditioning on observable characteristics. Average benefits are different from marginal benefits, and different policies select individuals at different margins. Conventional econometric methods that only account for selection on observables, like matching (:cite:`CocRub72`, :cite:`RoRu1983`, :cite:`HeIcSmTo98`), are not able to identify any parameter of interest (:cite:`HecVyr05`, :cite:`HeUrVy06`). For example, Carneiro (2011) (:cite:`Carneiro2011`) present evidence on agents selecting their level of education based on their unobservable gains and demonstrate the importance of adjusting the estimation strategy to allow for this fact. Heckman and Schmierer (:cite:`Heckman2010`) propose a variety of tests for the the presence of essential heterogeneity.
 
 .. todo::
 
@@ -75,9 +75,9 @@ It is common to summarize the average benefits of treatment for different subset
 treated :math:`\left(B^{TT}\right)` or untreated :math:`\left(B^{TUT}\right)`.
 
   .. math::
-      &B^{ATE} & = E [Y_1 - Y_0]\\
-      &B^{TT} & = E [Y_1 - Y_0 | D = 1]\\
-      &B^{TUT} & = E [Y_1 - Y_0 | D = 0]\\
+       B^{ATE} & = E [Y_1 - Y_0]\\
+       B^{TT} & = E [Y_1 - Y_0 | D = 1]\\
+       B^{TUT} & = E [Y_1 - Y_0 | D = 0]\\
 
 All average effect parameter possibly hide considerable treatment effect heterogeneity. The relationship between these parameters depends on the assignment mechanism that matches agents to treatment. If agents select their treatment status based on their own benefits, then agents that take up treatment benefit more than those that do not and thus :math:`B^{TT}` > :math:`B^{ATE}`. If agents select their treatment status at random, then all parameters are equal.
 

diff --git a/docs/figures/Join_Chat.svg b/docs/figures/Join_Chat.svg
diff --git a/docs/figures/fig-local-average-treatment.png b/docs/figures/fig-local-average-treatment.png
diff --git a/docs/figures/fig-weights-marginal-effect.png b/docs/figures/fig-weights-marginal-effect.png
diff --git a/docs/figures/scripts/resources/data.grmpy.info b/docs/figures/scripts/resources/data.grmpy.info
@@ -0,0 +1,77 @@
+
+
+ Number of Observations 
+
+                            Count
+
+  All                        1000
+  Treated                     811
+  Untreated                   189
+
+
+ Distribution of Outcomes
+
+                             Mean             Std-Dev.                  25%                  50%                  75%
+
+  All                      0.7240               3.2606              -1.3393               0.8122               2.9086
+  Treated                  1.0857               3.1382              -0.8966               1.0231               3.1695
+  Untreated               -0.8279               3.3282              -3.1614              -0.9844               1.5261
+
+
+ Distribution of Effects
+
+                             Mean             Std-Dev.                  25%                  50%                  75%
+
+  All                      2.9461               3.6042               0.4234               3.0131               5.4112
+  Treated                  2.8698               3.6111               0.2947               2.8721               5.3835
+  Untreated                3.2733               3.5655               0.8911               3.5938               5.5399
+
+
+ Criterion Function 
+
+  Value           2.681048804019      
+
+
+
+ MTE Information 
+
+    Quantile                Value
+
+          1%               1.8088
+          5%               2.1489
+         10%               2.3301
+         15%               2.4525
+         20%               2.5497
+         25%               2.6331
+         30%               2.7080
+         35%               2.7774
+         40%               2.8432
+         45%               2.9069
+         50%               2.9696
+         55%               3.0323
+         60%               3.0961
+         65%               3.1619
+         70%               3.2313
+         75%               3.3062
+         80%               3.3896
+         85%               3.4868
+         90%               3.6091
+         95%               3.7904
+
+
+ Parameterization 
+
+  Identifier                Value
+
+           0               1.2303
+           1              -2.3070
+           2              -1.7192
+           3              -1.6871
+           4               2.4869
+           5              -0.3900
+           6               2.1559
+           7              -0.2055
+           8               2.1975
+           9               2.8355
+          10               1.6985
+          11               2.8874
diff --git a/docs/figures/scripts/resources/data.grmpy.pkl b/docs/figures/scripts/resources/data.grmpy.pkl