User Julian Schuessler

38 votes

Accepted

A layman understanding of the difference between back-door and front-door adjustment

answered Jun 8, 2018 at 12:15

15 votes

Accepted

Representing interaction effects in directed acyclic graphs

answered Jul 21, 2015 at 11:00

9 votes

Accepted

Reference Request: Generalized Linear Models

answered Apr 20, 2014 at 15:29

8 votes

Accepted

Doubt in the Invariance Property of Consistent Estimators

answered Feb 13, 2021 at 8:23

7 votes

Is it possible to have a variable that acts as both an effect modifier and a confounder?

answered Jan 31, 2016 at 15:14

6 votes

Including Collider Variables in Prediction

answered Mar 28, 2019 at 15:29

6 votes

Accepted

Does the Heckman correction with an exclusion restriction provide causal inference?

answered Apr 10, 2019 at 14:01

6 votes

Accepted

If two sample means of $X$ differ significantly and $X$ predicts $Y$, can it be inferred that the samples differ in their $Y$?

answered Aug 9, 2020 at 14:45

6 votes

Accepted

Can we always write a random variable as conditional expectation plus error?

answered Feb 11, 2021 at 16:56

5 votes

Is there a difference between a causal relationship and a DIRECT causal relationship?

answered Apr 1, 2017 at 18:02

4 votes

Difference-in-difference model with mediators: Estimating the effect of different elements of a policy

answered Feb 20, 2017 at 10:27

4 votes

Would it be reasonable to attempt to use a propensity score as an instrumental variable?

answered May 27, 2017 at 14:01

4 votes

Accepted

Causal Inference - Revenue Per User

answered Jun 29, 2018 at 13:51

4 votes

Discrete choice panel models in R

answered Apr 26, 2015 at 12:50

4 votes

Accepted

Logistic regression with probability of success close to one

answered Mar 1, 2014 at 12:37

4 votes

Instrumental variables: In which cases would the average treatment effect on the treated (ATT) and local average treatment effect (LATE) be similar?

answered May 14, 2020 at 14:31

3 votes

Why does conditioning on a mediator variable makes the ancestor and the child independent?

answered May 20, 2020 at 9:07

3 votes

Accepted

How are standard exogeneity assumptions and indepent of potential outcomes concepts linked?

answered Aug 20, 2020 at 11:27

3 votes

Accepted

In observational studies, how can unconfoundedness, $(Y(1), Y(0)) \perp T \mid X$, hold if $(Y(1), Y(0))$ are fixed and non-random?

answered Oct 23, 2018 at 7:17

3 votes

Accepted

Synthetic control and unobserved confounders

answered Jan 23, 2019 at 13:54

3 votes

Is Fisher Sharp Null Hypothesis testable?

answered Jun 3, 2017 at 12:39

3 votes

Accepted

Are there any circumstances where adjusting for path variables / mediators is useful?

answered Feb 26, 2018 at 12:30

2 votes

Accepted

Directed Acyclic Graphs and the no unrepresented prior common causes assumption

answered May 17, 2018 at 13:56

2 votes

Accepted

Understanding the perfect randomization assumption in treatment models

answered Oct 14, 2018 at 12:39

2 votes

For the Average Treatment Effect (ATE) in causal inference, defined as $E(Y(1) - Y(0))$, what is $E(Y(1))$ usually referred to as?

answered May 24, 2019 at 15:44

2 votes

What to do with missing values?

answered Apr 20, 2014 at 15:14

2 votes

What are key papers discussing causal inference from a missing data perspective?

answered Jul 20, 2016 at 17:57

2 votes

How do we know if using IV model is more efficient than OLS?

answered Nov 19, 2015 at 15:20

2 votes

Textbooks/readings on what to do when you can't create an ideal experiment?

answered Nov 21, 2015 at 11:04

2 votes

Identifying a confounder

answered Jan 31, 2016 at 14:47

Stack Exchange Network

47 Answers

A layman understanding of the difference between back-door and front-door adjustment

Representing interaction effects in directed acyclic graphs

Reference Request: Generalized Linear Models

Doubt in the Invariance Property of Consistent Estimators

Is it possible to have a variable that acts as both an effect modifier and a confounder?

Including Collider Variables in Prediction

Does the Heckman correction with an exclusion restriction provide causal inference?

If two sample means of $X$ differ significantly and $X$ predicts $Y$, can it be inferred that the samples differ in their $Y$?

Can we always write a random variable as conditional expectation plus error?

Is there a difference between a causal relationship and a DIRECT causal relationship?

Difference-in-difference model with mediators: Estimating the effect of different elements of a policy

Would it be reasonable to attempt to use a propensity score as an instrumental variable?

Causal Inference - Revenue Per User

Discrete choice panel models in R

Logistic regression with probability of success close to one

Instrumental variables: In which cases would the average treatment effect on the treated (ATT) and local average treatment effect (LATE) be similar?

Why does conditioning on a mediator variable makes the ancestor and the child independent?

How are standard exogeneity assumptions and indepent of potential outcomes concepts linked?

In observational studies, how can unconfoundedness, $(Y(1), Y(0)) \perp T \mid X$, hold if $(Y(1), Y(0))$ are fixed and non-random?

Synthetic control and unobserved confounders

Is Fisher Sharp Null Hypothesis testable?

Are there any circumstances where adjusting for path variables / mediators is useful?

Directed Acyclic Graphs and the no unrepresented prior common causes assumption

Understanding the perfect randomization assumption in treatment models

For the Average Treatment Effect (ATE) in causal inference, defined as $E(Y(1) - Y(0))$, what is $E(Y(1))$ usually referred to as?

What to do with missing values?

What are key papers discussing causal inference from a missing data perspective?

How do we know if using IV model is more efficient than OLS?

Textbooks/readings on what to do when you can't create an ideal experiment?

Identifying a confounder