Skip to content
This repository has been archived by the owner on Jul 29, 2024. It is now read-only.

Changes to the expected value Calculation #18

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

sushmit86
Copy link

Proposing change
.. math:: E[Y^{*}] = P(Y | W=1) - P(Y| W=0),
Since Y is a binary Random variable with possible it would be more appropriate to change it to

.. math:: E[Y^{*}] = P(Y =1 | W=1) - P(Y =1 | W=0),

Proposing change 
.. math:: E[Y^{*}] = P(Y  | W=1) - P(Y| W=0),
Since Y is a binary Random variable with possible it would be more appropriate to change it to 

.. math:: E[Y^{*}] = P(Y =1 | W=1) - P(Y =1 | W=0),
@sushmit86
Copy link
Author

sushmit86 commented Apr 29, 2021

Also IMO it might be nice to add the derivation of the above using the Law of Total Expectation and the fact that E(Y|W=1) = P (Y=1|W=1) Using the fact that for Indicator Random Variables E(Y) = P(Y=1)

@shaddyab
Copy link
Contributor

In theory, Y can also be continuous.

@sushmit86
Copy link
Author

sushmit86 commented Apr 29, 2021

if Y is continuous how does the derivation work

E[Y^{*}] = P(Y | W=1) - P(Y| W=0)

What I have is this

E[Y_star] = E (Y * (W - p)/p * (1-p))
= E(Y * (1-p)/p*(1-p)| W=1)* P(W=1) + E(Y * (0 -p)/p(1-p)|W=0) * P(W=0) using Law of Total expectation
Simplifying the above we have

= E(Y|W=1) - E(Y|W=0)
if Y is continuous how are we arriving at the derived result

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants