chelsea's changes (#18)

ronikobrosly · Aug 20, 2020 · e1a1e3d · e1a1e3d
1 parent 93f9219
commit e1a1e3d
Show file tree

Hide file tree

Showing 7 changed files with 99 additions and 125 deletions.
diff --git a/causal_curve/mediation.py b/causal_curve/mediation.py
@@ -508,18 +508,21 @@ def calculate_mediation(self, ci=0.95):
 
         bootstrap_overall_means = np.array(bootstrap_overall_means)
 
-        final_results = (
-            pd.DataFrame(
-                {
-                    "Treatment_Value": self.t_bin_means,
-                    "Proportion_Direct_Effect": self.prop_direct_list,
-                    "Proportion_Indirect_Effect": self.prop_indirect_list,
-                }
-            )
-            .round(4)
-            .clip(lower=0, upper=1.0)
+        final_results = pd.DataFrame(
+            {
+                "Treatment_Value": self.t_bin_means,
+                "Proportion_Direct_Effect": self.prop_direct_list,
+                "Proportion_Indirect_Effect": self.prop_indirect_list,
+            }
+        ).round(4)
+
+        # Clip Proportion_Direct_Effect and Proportion_Indirect_Effect
+        final_results["Proportion_Direct_Effect"].clip(lower=0, upper=1.0, inplace=True)
+        final_results["Proportion_Indirect_Effect"].clip(
+            lower=0, upper=1.0, inplace=True
         )
 
+        # Calculate overall, mean, indirect effect
         total_prop_mean = round(np.array(self.prop_indirect_list).mean(), 4)
         total_prop_lower = self._clip_negatives(
             round(np.percentile(bootstrap_overall_means, q=lower * 100), 4)

diff --git a/docs/Mediation_example.rst b/docs/Mediation_example.rst
@@ -18,7 +18,7 @@ and mortality (Havumaki et al.):
 At some point though, it's helpful to validate these ideas with empirical tests.
 This tool provides a test that can estimate the amount of mediation that occurs between
 a treatment, a purported mediator, and an outcome. In keeping with the causal curve theme,
-this tool uses a test developed my Imai et al. when handling a continuous treatment and
+this tool uses a test developed by Imai et al. when handling a continuous treatment and
 mediator.
 
 In this example we use the following simulated data, and assume that the `mediator`

diff --git a/docs/changelog.rst b/docs/changelog.rst
@@ -4,6 +4,14 @@
 Change Log
 ==========
 
+Version 0.3.6
+-------------
+- Fixed bug in Mediation.calculate_mediation that would clip treatments < 0 or > 1
+- Fixed incorrect horizontal axis labels in lead example
+- Fixed typos in documentation
+- Added links to resources so users could learn more about causal inference theory
+
+
 Version 0.3.5
 -------------
 - Re-organized documentation

diff --git a/docs/conf.py b/docs/conf.py
@@ -22,7 +22,7 @@
 author = 'Roni Kobrosly'
 
 # The full version, including alpha/beta/rc tags
-release = '0.3.5'
+release = '0.3.6'
 
 # -- General configuration ---------------------------------------------------
 

diff --git a/docs/intro.rst b/docs/intro.rst
@@ -4,13 +4,12 @@
 Introduction to causal-curve
 ============================
 
-In academia and industry, randomized controlled experiments (or simply experiments or colloquially
-known as "A/B tests") are considered the gold standard approach for assessing the true, causal impact
+In academia and industry, randomized controlled experiments (or simply experiments or "A/B tests") are considered the gold standard approach for assessing the true, causal impact
 of a treatment or intervention. For example:
 
-* We want to increase the number of times per day new customers log into our business's website. Will it help if send daily emails out to our customers? We take a group of 2000 new business customers and half is randomly chosen to receive daily emails while the other half receives one email per week. We follow both groups forward in time for a month compare each group's average number of logins per day.
+* We want to increase the number of times per day new customers log into our business's website. Will it help if we send daily emails out to our customers? We take a group of 2000 new business customers and half is randomly chosen to receive daily emails while the other half receives one email per week. We follow both groups forward in time for a month compare each group's average number of logins per day.
 
-However, for ethical or financial reasons experiments may not always be feasible to carry out.
+However, for ethical or financial reasons, experiments may not always be feasible to carry out.
 
 * It's not ethical to randomly assign some people to receive a possible carcinogen in pill form while others receive a sugar pill, and then see which group is more likely to develop cancer.
 * It's not feasible to increase the household incomes of some New York neighborhoods, while leaving others unchanged to see if changing a neighborhood's income inequality would improve the local crime rate.
@@ -67,6 +66,16 @@ None of the methods provided in causal-curve rely on inference via instrumental
 rely on the data from the observed treatment, confounders, and the outcome of interest (like the above GPS example).
 
 
+Additional resources
+--------------------------------------------
+
+There are a number of excellent blogs and books covering the math behind causal inference in more depth.
+In addition to these, the following resources are recommended:
+
+- Miguel Hernan and Jamie Robin's book `"Causal Inference: What if" <https://cdn1.sph.harvard.edu/wp-content/uploads/sites/1268/2020/07/ci_hernanrobins_31july20.pdf>`_.
+- Cloudera Fast Forward Labs Research's `book on causal inference <https://ff13.fastforwardlabs.com/>`_.
+
+
 References
 ----------
 

diff --git a/examples/NHANES_BLL_example.ipynb b/examples/NHANES_BLL_example.ipynb
diff --git a/setup.py b/setup.py
@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="causal-curve",
-    version="0.3.5",
+    version="0.3.6",
     author="Roni Kobrosly",
     author_email="[email protected]",
     description="A python library with tools to perform causal inference using \