Causal Methods

"Causality. There is no escape from it, we are forever slaves to it. Our only hope, our only peace is to understand it, to understand the why." - Merovingian

[Causal Methods @causal_methods]

For those concerned about the validity of RCTs,

"A review of generalizability and transportability" - @IDegtiar and @sherrirose (2023) - https://t.co/YeYXwQYK9e #causalinference

[Causal Methods @causal_methods]

Another important reference: Degtiar and Rose (2023) - https://t.co/YeYXwQZhYM

[Causal Methods @causal_methods]

Wrt current twitter convos on generalizability of RCT results - Stuart (2015) has some relevant points - https://t.co/kkGkzImd7M

[Causal Methods @causal_methods]

Paper recos needed for graduate course reading on:

Dealing with noncompliance in survival analysis of clinical trials.

[Causal Methods @causal_methods]

Modeling the compliance process (noncompliance in RCT) should be standard practice. This necessitates collecting data on agents' ex-ante information set relevant to the compliance choice. Studying the choice environment is crucial for external validity (transportability).

[Causal Methods @causal_methods]

Hypothetical interventions should be read in conjunction with Frisch's causality as a mental exercise supposition and its elaboration in Pinto and Heckman's econometric causality. #EconTwitter

[Causal Methods @causal_methods]

Decision-theoretic foundations for statistical causality
- Dawid (2021) #decisionTheory #causality - https://t.co/cjNGltV2Ub

- foundations of decision-theoretic statistical causality
- accommodating stochastic variables
- intention to treat vs. intervention to treat

[Causal Methods @causal_methods]

Modern Dimension Reduction - Wagonner (2021) Elements series @CambridgeCore #causalinference - https://t.co/eSc1knqhrQ

- Open access till July 28th
- PCA, locally linear embedding, (t-SNE), unif. manifold approx. and projection, self-organizing maps & deep autoencoders
- R code

[Causal Methods @causal_methods]

Principles of Confounder Selection - Vanderweele (2019) - https://t.co/dqNfQKlUrF #causalinference

[Causal Methods @causal_methods]

Similar to model cards for ML model reporting, causal inference literature needs a template. It would greatly help in developing public knowledge graphs. What a resource that would be for researchers and practitioners!

https://t.co/9FA3luPMrZ

[Causal Methods @causal_methods]

Excellent set of tutorial papers from Statistics in Medicine - https://t.co/yIXeE8zQtG

[Causal Methods @causal_methods]

@totteh Would it be possible to add a reference journal article?

[Causal Methods @causal_methods]

Is it possible to teach causal inference frameworks/concepts/tools/insights to individuals in a way that helps them identify conspiracy theories early on? #causaltwitter

[Causal Methods @causal_methods]

Why is the potential outcomes framework dominant in Economics than DAGs? #causalinference

From Imbens (2020)

[Causal Methods @causal_methods]

To estimate the causal effect of hypothetical interventions to a system of factors represented as a DAG (observational data) - Castelletti and Consonni (2020) - https://t.co/UVzgXianUJ #causalinference

[Causal Methods @causal_methods]

Few recently read #DAG papers:
- Potential Outcomes vs. DAG - Imbens (2020) - https://t.co/TUyy7XoQJl
- Monotonicity in DAGs - Vanderweele and Robins (2009a) - https://t.co/1J4zF7Y3Xw

#causalinference #bayesnets #Epidemiology

[Causal Methods @causal_methods]

- DAG and other causal theories in Epidem.- Krieger and Smith (2016) - https://t.co/KAEe4kRvmo

[Causal Methods @causal_methods]

https://t.co/POjDfrOccv

[Niki Kilbertus @k__niki]

Excited about our @icmlconf workshop on the *Neglected Assumptions in Causal Inference* (https://t.co/XEUAzvblfH) with my amazing co-organizers @LauraBBalzer @alexdamour @uhlily @raziehnabi @ShalitUri

We warmly welcome contributions from outside CS as well (see CFP) @ICML2021

[Causal Methods @causal_methods]

Collection of papers linking categorical theory and Bayesian causal inference by @bgavran3 - https://t.co/Mz7BxoXVf1

#causalinference #bayesnet

[Causal Methods @causal_methods]

https://t.co/Q6kr0JuE0E

[Judea Pearl @yudapearl]

Good news! My lecture on "What is Causal Inference" is already available on youtube, free for all: https://t.co/RFRxUXnAAd
Enjoy!
And, if there are questions, do not hesitate, preferably after reading #Bookofwhy, for the obvious reason that the answer is very likely to be there.