finished mediation documenation, fixed mediation bug

causal_curve/mediation.py

@@ -7,9 +7,8 @@
 
 import numpy as np
 import pandas as pd
-from pandas.api.types import is_float_dtype, is_numeric_dtype
+from pandas.api.types import is_float_dtype
 from pygam import LinearGAM, s
-from tqdm import tqdm
 
 from causal_curve.core import Core
 from causal_curve.utils import rand_seed_wrapper
@@ -438,8 +437,8 @@ def calculate_mediation(self, ci=0.95):
         Parameters
         ----------
         ci: float (default = 0.95)
-            The desired bootstrap confidence interval to produce. Default value is 0.95, corresponding
-            to 95% confidence intervals. bounded (0, 1.0).
+            The desired bootstrap confidence interval to produce. Default value is 0.95,
+            corresponding to 95% confidence intervals. bounded (0, 1.0).
 
         Returns
         ----------
@@ -532,7 +531,9 @@ def calculate_mediation(self, ci=0.95):
         )
 
         print(
-            f"\n\nMean indirect effect proportion: {total_prop_mean} ({total_prop_lower} - {total_prop_upper})"
+            f"""\n\nMean indirect effect proportion:
+            {total_prop_mean} ({total_prop_lower} - {total_prop_upper})
+            """
         )
         return final_results
 

docs/Mediation_example.rst

@@ -87,13 +87,17 @@ around 52% (+/- 1%) of the effect of the treatment on the outcome. This indicate
 moderate mediation is occurring here. The remaining 48% occurs through a direct effect of the
 treatment on the outcome.
 
-The report also shows how this mediation effect various as a function of the continuous treatment.
-In this case, it looks the effect is relatively flat (as expected).
-
 So long as we are confident that the mediator doesn't play another role in the causal graph
 (it isn't a confounder of the treatment and outcome association), this supports the idea that
 the mediator is in fact a mediator.
 
+The report also shows how this mediation effect various as a function of the continuous treatment.
+In this case, it looks the effect is relatively flat (as expected). With a little processing
+and some basic interpolation, we can plot this mediation effect:
+
+.. image:: ../imgs/mediation/mediation_effect.png
+
+
 
 References
 ----------

docs/changelog.rst

@@ -4,6 +4,12 @@
 Change Log
 ==========
 
+Version 0.2.1
+-------------
+- Fixed Mediation tool error / removed `tqdm` from requirements
+- Misc documentation cleanup / revisions
+
+
 Version 0.2.0
 -------------
 - Added new Mediation class

docs/conf.py

@@ -22,7 +22,7 @@
 author = 'Roni Kobrosly'
 
 # The full version, including alpha/beta/rc tags
-release = '0.2.0'
+release = '0.2.1'
 
 
 # -- General configuration ---------------------------------------------------

docs/contribute.rst

@@ -44,7 +44,7 @@ Feature request
 A good feature request usually contains:
 
 - a description of the requested feature,
-- a description of the relevance of this feature to time series classification,
+- a description of the relevance of this feature to causal inference,
 - references if applicable, with links to the papers if they are in open access.
 
 This makes reviewing the relevance of the requested feature easier.

docs/index.rst

@@ -40,7 +40,7 @@ Welcome to causal-curve's documentation!
    :caption: Contents:
 
 
-**causal-curve** is a Python package dedicated with tools to perform causal inference
+**causal-curve** is a Python package with tools to perform causal inference
 using observational data when the treatment of interest is continuous.
 
 
@@ -72,7 +72,7 @@ This library attempts to address this gap, providing tools to estimate causal cu
 
 
 Quick example (of the ``GPS`` tool)
---------------------------------------
+-----------------------------------
 
 **causal-curve** uses a sklearn-like API that should feel familiar to python machine learning users.
 The following example estimates the causal dose-response curve (CDRC) by calculating
@@ -105,18 +105,30 @@ generalized propensity scores.
 
 6. Explore or plot your results!
 
+None of the methods provided in causal-curve rely on inference via instrumental variables, they only
+rely on the data from the observed treatment, confounders, and the outcome of interest (like the above GPS example).
 
-`Getting started <install.html>`_
----------------------------------
 
-Information to install, test, and contribute to the package.
 
+A caution about assumptions
+---------------------------
+
+There is a well-documented set of assumptions one must make to infer causal effects from
+observational data. These are covered elsewhere in more detail, but briefly:
+
+- Causes always occur before effects: The treatment variable needs to have occurred before the outcome.
+- SUTVA: The treatment status of a given individual does not affect the potential outcomes of any other individuals.
+- Positivity: Any individual has a positive probability of receiving all values of the treatment variable.
+- Ignorability: All major confounding variables are included in the data you provide.
 
+Violations of these assumptions will lead to biased results and incorrect conclusions!
 
+In addition, any covariates that are included in `causal-curve` models are assumed to only
+be **confounding** variables.
 
-Indices and tables
-==================
 
-* :ref:`genindex`
-* :ref:`modindex`
-* :ref:`search`
+
+`Getting started <install.html>`_
+---------------------------------
+
+Information to install, test, and contribute to the package.

setup.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="causal-curve",
-    version="0.2.0",
+    version="0.2.1",
     author="Roni Kobrosly",
     author_email="[email protected]",
     description="A python library with tools to perform causal inference using \