Questions tagged [causality]
The relationship between cause and effect.
0
votes
0answers
26 views
IPW for the effect of treatment on treated with a continuous treatment
I've been banging my head against the wall trying to figure out how to construct inverse probability of treatment weights (IPTW) for the population average effect of treatment on treated (PATT) with a ...
1
vote
1answer
16 views
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?
For the Average Treatment Effect (ATE) in causal inference, is it usually defined as
$$
E(Y(1) - Y(0))
$$
I am wondering what the most commonly referred to name for $E(Y(1))$ is? Is it not the ...
1
vote
1answer
26 views
Does the t-test require randomization?
Can I use t test for a non-equivalent quasi-experimemtal design? As there is no randomization, can it violate the assumptions of the t-test? What statistical technique should I use?
0
votes
1answer
16 views
In the potential outcomes framework, does conditioning on the potential outcomes automatically imply knowledge of the treatment assignment?
Suppose we have that $Z\in \{0,1\}$ is the treatment, $(Y(1),Y(0))$ the potential outcomes, and $X$ the covariates. Suppose we have know that unconfoundedness holds on $X$, such that:
$$
(Y(1),Y(0)) \...
0
votes
0answers
29 views
Review paper on causal modelling of complex networks
Although I have a growing interest in network neuroscience and complex neuroscience in general, I have quite a bit of trouble following Twitter discussions on causal modelling of brain network ...
1
vote
0answers
11 views
In causal inference, is the usual unconfoundedness assumption interpreted to apply at the unit or covariate level?
Suppose for each unit $i \in \{1, \ldots, N\}$, we have that
$(Y_i(1),Y_i(0))$ are the potential outcomes,
$Z_i$ is the treatment,
and
$X_i$ the covariates.
I have seen the following two ...
0
votes
2answers
41 views
If unconfoundedness holds under a set of covariates $X$, will it also hold on an extended set of covariates, $(X,X')$?
Suppose unconfoundedness holds for a set of potential outcomes $(Y(1),Y(0))$ and treatment $Z$, conditional on a set of covariates, $X$ such that:
$$
(Y(1),Y(0)) \perp Z \mid X
$$
Then, is it ...
0
votes
1answer
28 views
In Exact Matching in Causal Inference, why is it that that $P(X_i=x\mid Z_i = 1)= P(X_i=x\mid Z_i = 0)$ where $X,Z$ are the covariates and treatment?
In Exact Matching in Causal Inference, I read that because we assume exact matches, then exact balance occurs in the distribution of the covariates. It is then often stated that if $X,Z$ are the ...
1
vote
1answer
15 views
Propensity score: which treatment effect is easier to infer?
I'm currently working on a study where the goal is to estimate the treatment effect of a binary exposure. I want to calculate the Average Treatment Effect (ATE), Average Treatment Effect in the ...
4
votes
0answers
25 views
conditional and interventional expectation
Conditional expectation $E[Y|X]$ and interventional expectation $E[Y|do(X)]$ are related but conceptually very different things.
I know that if $X$ is a randomly assigned by an experiment, we have ...
0
votes
1answer
82 views
Using Machine Learning to Estimate Causal Effects from Observational Data
I would like to use machine learning to predict a categorical (ordinal, multi-class) outcome variable from a cross-sectional dataset with about 20,000 observations and 300 features. Importantly, I ...
0
votes
1answer
15 views
What's the difference between a “surrogate metric” and a “proxy metric”?
Is there a difference between a "surrogate metric" a "proxy metric" or a "correlated metric"? Is "surrogate" simply unnecessary jargon, or is there a meaning that makes it more specific than when ...
2
votes
0answers
13 views
How to choose the control variables in the conditional expectation to hold fixed when studying a causal relationship
I'm reading the introductory chapter of the wooldridge's book, "Econometric analysis of cross section and panel data". The chapter begins by highlighting the role and importance of conditional ...
1
vote
1answer
36 views
Is there an R package that can infer a causal structure with a mix of discrete and continuous variables? [closed]
I have an observational dataset with a mixture of discrete and continuous variables. I'd like to infer a causal structure compatible with the data.
The pcalg package for R can handle datasets with ...
0
votes
0answers
13 views
Kernel-based Propensity Score Matching diff-in-diff
I want to perform the Kernel-based Propensity Score Matching diff-in-diff. I am actually using the following command. The diff-in-diff result for this code is 0.000. Please, would someone help in ...
3
votes
1answer
67 views
What does it mean to “non-parametrically” identify a causal effect within the super-population perspective in causal inference?
I am wondering, within the context of causal inference, what it means to "non-parametrically" identify a causal effect within the super-population perspective. For example, in Hernan/Robins Causal ...
0
votes
1answer
43 views
Matching with Multiple Treatments
What's the best way to use matching methods with multiple treatment groups?
I'm assessing the impact of an intervention on an outcome. For my first analysis, I used the MatchIt package (see code below)...
1
vote
1answer
26 views
In causal inference, why is the uncounfoundedness assumption interpreted as the treatment assignment being conditionally independent of the outcomes?
In causal inference, the unconfoundedness assumption is usually stated as:
$$
Y(1), Y(0) \perp Z \mid X
$$
or
$$
P(Z \mid Y(1), Y(0), X) = P(Z \mid X)
$$
where $Z$ is the treatment assignment, $(Y(1)...
0
votes
1answer
19 views
In the study of causal inference on networks, what implications does assuming unconfoundedness have on homophily?
I read in a book on causal inference on networks that:
The unconfoundedness assumption does not rule out the presence of
homophily, that is tendency of individuals who share similar
...
0
votes
1answer
29 views
Why does unconfoundedness not hold if the covariates are affected by the treatment assignment?
I found a mention inside the Rubin-Imbens Causal Inference book that:
Unconfoundedness is generally violated if X includes variables that
are themselves affected by the treatment. For example, in ...
0
votes
1answer
27 views
Why is the infinite-population perspective usually taken in sampling? What underlying bias/convergence implications are there?
In sampling literature and causal inference literature, there usually is a distinction made about how to view observed data. The first is usually to view some observed data as having come from a ...
1
vote
1answer
25 views
Difference between covariates and treatment confounders in propensity score matching
Here, I have the definition of a propensity score:
Propensity score is defined as the conditional probability of assignment to a treatment given a vector of covariates including the values of all ...
0
votes
1answer
16 views
Determining Conditional Independence from Marginal Independence?
so if I have a 3 columns of binary variables X, Y and Z with their respective values and I would like to determine whether X,Y are conditionally independent given Z. How can I go about doing this? ...
2
votes
2answers
27 views
Matching covariate selection- should we not match on binary features where overwhelming majority of subjects fall in one category?
I am trying to investigate if there is a relation between Occupational Therapy (OT) dosage for stroke patients and patient recovery. I have separated the patients into 2 groups by the amount of ...
3
votes
0answers
20 views
Before using CV-selected Regression model for Inference, shouldn't model performance be evaluated on unused test set?
I just came across a biokinesiology paper that used some Machine Learning methods, but I think there is a flaw in their methodology.
The authors had data on stroke patients and used Lasso regression ...
3
votes
1answer
28 views
How to estimate the “effect of an effect” identified through regression?
Suppose I estimate the effect of some variable $X_t$ on $Y_t$. Let's say that $X_t$ is some firm variable (age of CEO in year $t$ or something like that), and $Y_t$ is a firm's change in investment in ...
0
votes
0answers
20 views
Estimating Policy Effect with a Logit
So I am testing a policy which was introduced in a country trying to incentivise people to stay employed at older ages (beyond the retirement age of 65). As such, they introduced a bonus where people ...
4
votes
2answers
80 views
Why does Judea Pearl call his causal graphs Markovian?
In his texts on causality, Judea Pearl always refers to the simplest graphs he uses, i.e. the acyclic graphs with independent confounders, as Markovian. I don't see why these graphs contain anything ...
0
votes
0answers
17 views
Causal interpretation using first-stage residuals as controls in second stage regression
I have a main model that investigates the effect of X1 on an outcome variable Y, controlling for certain exogenous variables <...
2
votes
1answer
30 views
How do I apply weights to a Cox Regression Model in R?
I am trying to answer the question of whether service in a certain organization has an effect on age of first marriage, and am interested in using the Cox model
to understand the difference in the ...
1
vote
0answers
35 views
Ideas for research projects: Causality, DAGs etc [closed]
I am looking for ideas for research projects to give my students in a module on causality and DAGs (third-year undergraduate and master level).
I had a couple of ideas. For example, one could, for ...
4
votes
1answer
58 views
Front-door criteria - what does the second requirement mean?
I'm reading Causal Inference in Statistics: A Primer by Pearl, et. al. and I'm a little confused by the definition of front door. The definition in the book (Definition 3.4.1) is:
A set of ...
3
votes
1answer
40 views
IPTW for multiple treatments
I am dealing with a dataset where patients are subjected to multiple treatments A or B or C or D . Since there are four treatment options I am using multinomial regression to estimate the propensity ...
0
votes
0answers
11 views
Non-collapsibility issue in OR and mediation analysis with binary DV, binary mediator and 3-category IV
I have been looking for theory and software implementation to solve for non-collapsibility issue in mediation analysis with categorical DV and mediator (and IV). In my case, the DV and mediator are ...
1
vote
1answer
69 views
Propensity Scores: What is this estimator?
I'm reading The Central Role of the Propensity Score in Observational Studies for Causal Effects in order to understand why Propensity Scores work. I'm kind of new to this and I'm not understanding an ...
1
vote
2answers
294 views
Does the Heckman correction with an exclusion restriction provide causal inference?
I think I might be getting instrumental variables estimation and the Heckman correction with an exclusion restriction confused. I know that instrumental variables estimation is way to show causal ...
2
votes
1answer
62 views
Find the cause of periodic behaviour present in Google-search “war”
If you use Google Trends (a tool to illustrate the frequency of search terms) and search for the word "war" with USA as geographic region, you will be presented with the graph below showing monthly ...
1
vote
1answer
38 views
In matching procedures in causal inference, will matching on more covariates always decrease the bias? Can the bias ever increase? [duplicate]
There are various matching procedures in the causal inference literature, from exact matching to propensity score matching and more. The goal is usually to find the Average Treatment Effect (ATE). ...
1
vote
0answers
69 views
Structural equation and causal model in economics
Structural equations is a useful language for causal analysis in economics. In Causality Pearl (2009 cap 5) we can find the best discussion about this.
My question: is possible to use the concept of ...
2
votes
1answer
41 views
Including Collider Variables in Prediction
When the goal is to estimate a causal association between X and Y in the regression framework, one should not condition on (include as covariates) collider variables (common causes of both X and Y) ...
0
votes
0answers
18 views
How to deal with circular causality
Often in time series and panels, the "dependent" and "causal" variable don't share purely that relationship. There is a fair bit of reverse causality as well. ,e.g.
x causes y, but then either y ...
1
vote
0answers
37 views
AIC for Causal Inference
I read a post explaining why the Akaike Criterion cannot be used for deciding if A cause B or B caused A.
I'm curious about a more general case of using AIC for causal inference (with observational ...
0
votes
0answers
37 views
Causality logic appears reversed. What's the explanation?
In reading Detecting Causality in Complex Ecosystems I came across the following passage:
Our alternative approach [...] tests for causation by measuring the extent to which the historical record ...
1
vote
1answer
37 views
How to figure out whether a coin is “weighted” with some number of flips
Suppose there's a weighted coin. That coin either lands on heads every 1/10 times it is tossed, or it never lands on heads at all. I don't know whether that coin is the type of coin that lands on ...
0
votes
0answers
43 views
Proof of Berkson's Paradox
I'd be very thankful if someone could help me with the proof of Berkson's Paradox. I found this quite helpful thread which I understand How to prove Berkson's Fallacy?.
But I'm actually trying to ...
1
vote
0answers
44 views
If a study indicates no correlation between two variables, does it also indicate a lack of casual relationship? [duplicate]
Of course, correlation does not equal causation. But I am having trouble understanding if there is no correlation between two variables, would this indicate a lack of casual relationship between them ...
2
votes
1answer
59 views
Relationship between Causal Calculus (in the sense of the Book of Why) and other existing modeling formalisms?
I am watching this video on youtube: https://youtu.be/zvrcyqcN9Wo?t=2896
about Causal Calculus (CC), namely this section on causal graphs, and it seems to me that this theory of causality is not that ...
1
vote
1answer
39 views
Example distribution to match an example?
I am doing exercises from "Causal Inference in Statistics: A Primer", by Pearl et al (2016). In chapter 1.2 there is a training challenge that goes like:
In an attempt to estimate the effectiveness ...
0
votes
0answers
75 views
Term for two variables that are “too close for control”
Sometimes we are tempted to assess a relationship of X1 with Y while controlling for X2, but it would be a mistake, because X2 is not merely correlated with Y -- it is more closely associated than ...
0
votes
0answers
25 views
Instrumental variable predicts endogenous variable in an unexpected direction
I am trying to estimate a causal relationship between two variables. I'm concerned about endogeneity, so I'm using an instrument. The instrument strongly predicts the endogenous regressor, however it ...
