Skip to main content

Questions tagged [counterfactuals]

An *if* statement in which the condition is untrue or unrealized. Used in causal analysis for comparing potential outcomes under different hypothesized conditions.

Filter by
Sorted by
Tagged with
3 votes
1 answer
22 views

How to interpret the conditional expectation mean $E[Y(0)\mid Z=1]$ where $Y(0)$ is the potential outcome on control and $Z$ the treatment assignment?

How can I interpret the conditional expectation mean $E[Y(0)\mid Z=1]$ where $Y(0)$ is the potential outcome on control and $Z$ the treatment assignment? I believe it is the "expected outcome if ...
user321627's user avatar
  • 4,448
1 vote
1 answer
167 views

Average treatment effect: counterfactual and graphical derivation

I have some (shameful) doubts about the Average Treatment Effect (ATE), also known as Average Causal Effect (ACE). In this setting, I am interested in a binary exposure/treatment variable ...
wrong_path's user avatar
1 vote
0 answers
38 views

Impact of counterfactual hiring decisions

As the hiring manager for my widget company, in order to decide whether to progress an applicant to an interview I administer an IQ test. If the applicant's score is above 110 on the test I let them ...
quant's user avatar
  • 384
0 votes
0 answers
27 views

How can I compute a counterfactual query for discrete binary variables?

In computing counterfactual queries for structural causal models (SCMs), J. Pearl says to go through 3 steps. Abduction Action Prediction In his book, Causal Inference in Statistics, he shows how to ...
Jane Wayne's user avatar
  • 1,399
4 votes
1 answer
48 views

AB Testing with and without counterfactuals

My boss wants us to run a simple experiment where we manipulate the search results on our website. He is insisting that we must have counterfactuals for the results to be valid. The way I see ...
Daniel's user avatar
  • 221
3 votes
2 answers
133 views

Does Pr(Y | X=x) equal Pr(Y | do(X=x)) in a randomized experiment?

On page 435 of Cosma Shalizi's advanced data analysis book (link: https://www.stat.cmu.edu/~cshalizi/ADAfaEPoV/ADAfaEPoV.pdf), he states the following about randomized experiment for causal inference: ...
VDCN's user avatar
  • 31
0 votes
0 answers
21 views

Significance of pie-charts in representing sufficient-cause components

In the book Causal Inference - What If, the authors present a form of pie-chart-type diagram to represent sufficient-cause components (refer to figure 5.1 on page 64 and figure 5.2 on page 65 ...
Anirban Chakraborty's user avatar
3 votes
1 answer
266 views

Why have both standardization and IP weighting, when they are fundamentally similar?

In Causal Inference: What If by Hernan and Robins, I came across two methods to find causal effect - standardization (page 19) and IP weighting (page 20). It seems both methods are fundamentally same. ...
Anirban Chakraborty's user avatar
1 vote
0 answers
49 views

Variable selection in causal inference regression models when $p > n$?

Are there accepted techniques for selecting variables in causal inference (not prediction) where the number of variables exceeds our sample size, making a standard OLS regression impossible? Assume ...
RobertF's user avatar
  • 6,184
4 votes
1 answer
84 views

Creating deliberate missing data to impute a counter factual

This question is more out of curiosity than necessity. Any useful resources/literature will be massively appreciated. Recently started research in a new field that I’m unfamiliar with: economic policy ...
Rhys Maredudd Davies's user avatar
3 votes
0 answers
20 views

Clarification Needed on Derivation in Judea Pearl's "Detecting Latent Heterogeneity" (2015)

I'm reading a paper by Judea Pearl titled "Detecting Latent Heterogeneity" from 2015, which can be found at this link: Detecting Latent Heterogeneity I'm having trouble understanding a ...
snapper's user avatar
  • 131
1 vote
1 answer
61 views

Noise abduction for computing counterfactuals

Given observational data $X$ and knowledge of the true causal graph structure $\mathcal{G}$. How does abduction of the exogenous noise ($U$) for computing counterfactuals work? We don't have data ...
CausalQuestions's user avatar
1 vote
1 answer
162 views

Computing counterfactual query given an SCM and how it differs from computing interventional query?

Assuming we have the following structural causal model (SCM), with a confounder DAG structure, as follows: Noise variables: $$U_1 \sim \mathcal{N}(0,\,1)$$ $$U_2 \sim \mathcal{N}(0,\,1)$$ $$U_3 \sim \...
CausalQuestions's user avatar
3 votes
1 answer
37 views

Relevance of baseline covariates in the reference group when estimating ATT

Let's say we have a single treated group A, and then two reference groups B and C. The two ...
elbord77's user avatar
  • 557
3 votes
1 answer
104 views

Should pre-intervention outcome variable Y0 be excluded from counterfactual model based on causal diagram?

Based on the causal diagram below, where: Y1 is the outcome in the post-intervention period, Y0 is the pre-intervention outcome, X is a Yes/No healthcare intervention, and Z1 & Z2 represent ...
RobertF's user avatar
  • 6,184
5 votes
1 answer
76 views

Synthetic Controls - Comparing Two Treatments

All of the literature on synthetic control groups examines situations where 1 or more units receive a treatment and are compared to non-treated units. I am wondering if there is any statistical or ...
B_Miner's user avatar
  • 8,810
5 votes
1 answer
111 views

Representation of unconfoundedness of Rubin causal models on Pearl causal models

Disclaimer: I'm new to the potential outcome framework so this question may not make lot of sense, in case please let me understand where I am failing. I was reading this question about the ...
DaSim's user avatar
  • 460
8 votes
1 answer
375 views

How to think about counterfactuals?

I am trying to wrap my mind around causal inference, mostly because I think it will be useful for my work. I read the Book of Why, I read "Causal inference in statistics: A primer", I did an ...
January's user avatar
  • 7,637
3 votes
1 answer
77 views

Derivation of equation 7.6 in Pearl's Causailty (2nd ed): counterfactual given actual observations

Background: Pearl presents the following Equation (7.6) for the probability of a counterfactual, $Y_{x'} = y'$, given the observed quantities $X=x, Y=y$: $$ \begin{align} P(Y_{x'} = y'| X=x, Y=y) &...
einar's user avatar
  • 4,282
2 votes
1 answer
50 views

How to convert a Gamma GLM to a Pearl's Structural Causal Model (SCM) with exogenous noise terms?

Assume both X and Y are univariate for this. So in Judea Pearl's SCM, you have endogenous variables and exogenous noise. A regular causal Bayesian Network like X->Y where X and Y|X is normal can be ...
Vattaka's user avatar
  • 232
2 votes
0 answers
131 views

Question about using potential outcomes in DAGs in real world example

I am trying to understand how DAGs and potential outcomes look together. I came across these excellent posts (here and here, but I am trying to understand how this looks in a real world example. ...
BobC's user avatar
  • 39
1 vote
2 answers
147 views

In a diff-in-diff or regression discontinuity research design, why is it important to describe why the counterfactual is a plausible one?

I've heard it mentioned that in difference-in-differences, regression discontinuity, or even in some other quasi-experimental research designs, that the counterfactual should be explained as a ...
simplycoding's user avatar
2 votes
1 answer
155 views

Counterfactual Bayesian survival analysis in pymc

I am trying to determine mortality rates for untreated patients from an observational dataset where treatment has occurred (thus blocking the possibility of further untreated mortality). You can't ...
Mike's user avatar
  • 103
3 votes
0 answers
228 views

Matching on pre-treatment outcome $z$-score in diff-in-diff analysis to avoid regression toward the mean bias in $ATT$ estimates?

There have been many articles (e.g., Chabé-Ferret (2017), Daw & Hatfield (2018), Zeldow & Hatfield (2021)) discussing the perils of matching on pre-treatment outcomes (such as patient's ...
RobertF's user avatar
  • 6,184
4 votes
0 answers
420 views

Should inverse probability weighting be used in two-way fixed-effects panel regression?

Let's assume a (balanced) panel data set with two measurement points $t_0$ and $t_1$, where $t_0$ may be considered as the baseline. Some of the ID's are treated at $t_1$, i.e. $D=1$, the assignment ...
jay.sf's user avatar
  • 878
2 votes
0 answers
642 views

Conditional exchangeability when conditioning on continuous variable

Conditional exchangeability is often introduced in a simple setting with a binary outcome, binary treatment, and binary confounding variable. In this setting, exchangeability holds within strata of ...
Wissenschaft's user avatar
9 votes
4 answers
1k views

Textbook recommendations covering machine learning techniques for causal inference?

Over the past 15 years there has been progress in adapting machine learning methods for causal inference. For example: targeted learning, double machine learning, causal trees. Is there a textbook ...
RobertF's user avatar
  • 6,184
5 votes
0 answers
275 views

Counterfactual Estimation - Common Practices in Applied Causality

I am quite new to the topic and trying to figure out a workflow for causal analysis. My aim is to establish a baseline of ATE (I think) and then experiment with disentangled representations and ...
Nikos H.'s user avatar
1 vote
0 answers
211 views

Constrastive vs Counterfactual Explanations

Is there any canonical definition for Contrastive vs Counterfactual explanations? In the literature, I keep reading different versions but I wonder if there are good definitions or illustrative ...
Carlos Mougan's user avatar
1 vote
1 answer
526 views

Correct methodology using $g$-computation to estimate Average Treatment Effect on the Treated ($ATT$)?

I have a question about the $g$-computation methodology for estimating the Average Treatment Effect on the Treated ($ATT$) in the following article. The authors recommend estimating the $ATT$ by first ...
RobertF's user avatar
  • 6,184
1 vote
0 answers
297 views

Generalized DiD with non-binary treatment variable? How to proceed?

I have a certain law as my (non-binary) treatment variable (range: 0-2) at the state level which takes on different values based on whether it prohibits altogether (2), limits (1), or allows the ...
Ahmed Chaudhry's user avatar
0 votes
1 answer
193 views

The average treatment effect and the difference in means

Hi I have a question related to the treatment effect. Recently, I am reading literatures on treatment effect and have a question. In the literatures, we denote the counterfactual outcomes as $Y_1$ and ...
M.C. Park's user avatar
  • 935
2 votes
1 answer
115 views

Using counterfactual modeling techniques to assess racial bias in predictive models

My team at a health insurance company is discussing how we might measure racial bias in the various predictive models our company uses to assess future health risk (such as annual medical cost or ...
RobertF's user avatar
  • 6,184
4 votes
0 answers
1k views

Linear model: potential outcome framework vs. structural causal model

From my reading about the potential outcomes framework (POF) and structural causal models (SCM), I understand that both perspectives have been shown to be equivalent but take different starting points....
curious's user avatar
  • 98
3 votes
1 answer
310 views

find upper and lower bounds on average causal effect $\theta$

I am working on some problems in All of statistics by Wasserman, and I am not quite sure how to tackle this problem. Suppose you are given data $(X_1, Y_1), \dots, (X_n, Y_n)$ from an observational ...
MoneyBall's user avatar
  • 907
23 votes
3 answers
7k views

Strong ignorability: confusion on the relationship between outcomes and treatment

In the research area of potential outcomes and individual treatment effect (ITE) estimation, a common assumption called ''strong ignorability'' is often made. Given a graphical model with the ...
SolingerStuebchen's user avatar
2 votes
1 answer
566 views

Difference between the counterfactual mean and average treatment effect

I am new on the causality topic, I don't know the difference between the average treatment effect and counterfactual mean. Can anyone tell me?
cookie monster's user avatar
8 votes
1 answer
3k views

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

It seems that if the proportion of always-takers in the control group (to whom eligibility was not assigned) is much smaller than the proportion of compliers in the treatment group (to whom ...
Aqqqq's user avatar
  • 569
4 votes
1 answer
150 views

Adjustment formula for counterfactuals: can we get rid of $X=x$?

Pearl et al. "Causal Inference in Statistics: A Primer" (2016) p. 108 contains the following adjustment formula (based on the backdoor criterion) for probabilities of counterfactuals expressed using ...
Richard Hardy's user avatar
2 votes
1 answer
306 views

Counterfactual Expectation Calculation

$\newcommand{\doop}{\operatorname{do}}$ Problem: (This is from Study question 4.3.1 from Causal Inference in Statistics: A Primer, by Pearl, Glymour, and Jewell.) Consider the causal model in the ...
Adrian Keister's user avatar
3 votes
1 answer
106 views

How can I update a disease prediction model with new treatment group data while maintaining the original causal relationships?

Context: I have a prediction model which predicts the probability of getting a disease. This prediction model has been created based on data of patients who did not get any form of treatment. I use ...
Koot6133's user avatar
4 votes
1 answer
910 views

How to understand probability of Necessity (PN) ≥ 100%, as in this example from 'Causal Inference in Statistics a primer'

In the book 'Causal Inference in Statistics A Primer' By Pearl et al. there is an example towards the end, (Ex 4.5.1 page 119) that calculates the probability of necessity PN = 1, and the authors ...
Mint's user avatar
  • 141
4 votes
1 answer
136 views

Counterfactuals in Econometric Modeling (Abortion-Crime Hypothesis Revisited)

Donohue and Levitt (2019) recently published a working paper revisiting the abortion-crime link. My question is specific to equation (2) in their paper (see below): $$ ln(CRIME_{st}) = \beta_{1}...
Thomas Bilach's user avatar
1 vote
0 answers
24 views

What are counterfactual objectives?

What is a simple-to-understand definition/explanation for counterfactual objectives in Machine Learning? I heard this term in this 3 minutes paper presentation https://youtu.be/TOpJ4g9v-H4.
Andrei Margeloiu's user avatar
2 votes
1 answer
556 views

Offline evaluation of Counter factual data for Recommendation

I am building new model and facing at offline evaluation tasks. My goal is to predict higher CTR(=click/impression) advertisement, and improve sales.(sales would improve if user watch more ...
Koji Sugano's user avatar
6 votes
2 answers
383 views

The counterfactual model for causation

My professional training took place in the late 1990's and I don't recall hearing some of the terminology that seems nigh-universal nowadays. I believe the accepted name is "counterfactual model for ...
Brent Hutto's user avatar
  • 1,098
5 votes
1 answer
138 views

Conterfactual estimation in machine learning model

There are various techniques to build counterfactual estimations of certain variables for linear models in observational studies. Some of those are based on comparing the change in the predicted ...
Bakaburg's user avatar
  • 2,939