explain purple in overlap plots

Signed-off-by: Nathaniel <[email protected]>

Author: NathanielF

docs/source/notebooks/inv_prop_pymc.ipynb

@@ -791,7 +791,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Here we have plotted  in three panels: (1) mirrored draws from the propensity score distribution split by treated and control groups (2) the expected outcome in those groups under re-weighting under each draw (3) the derived estimates for the average treatment effect. Note here how expected value of the ATE is pulled slightly away from the true value under this weighting scheme. This is likely due to the high number of individuals with extreme propensity scores - denoted in (1) as individuals with propensity scores in excess of .9 and below .1.\n",
+    "Here we have plotted  in three panels: (1) mirrored draws from the propensity score distribution split by treated and control groups in the red and blue with the purple showing the pseudo-population created by the weighting (2) the expected outcome in those groups under re-weighting under each draw (3) the derived estimates for the average treatment effect. Note here how expected value of the ATE is pulled slightly away from the true value under this weighting scheme. This is likely due to the high number of individuals with extreme propensity scores - denoted in (1) as individuals with propensity scores in excess of .9 and below .1.\n",
     "\n",
     "Let's check what happens using the overlap weighting scheme?"
    ]
@@ -820,7 +820,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We see here how the particular weighting scheme was able to recover the true treatment effect. This is a useful reminder in that, while propensity score weighting methods are aids to inference in observational data. Not all weighting schemes are created equal and we need to be careful in our assessment of when each is applied appropriately. "
+    "We see here how the particular weighting scheme was able to recover the true treatment effect by defining a contrast in a different pseudo population. This is a useful reminder in that, while propensity score weighting methods are aids to inference in observational data. Not all weighting schemes are created equal and we need to be careful in our assessment of when each is applied appropriately. "
    ]
   },
   {