Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Causal inference as a blind spot of data scientists

dzidas.com

Post reply on HN
Print view View original

81–90 of 106 posts

Re: Causal inference as a blind spot of data scientists

#81

Post by spywaregorilla »

morelandjs wrote: post #71
spywaregorilla wrote: post #68

> The DoubleML method is founded on machine learning modeling and consists of two key steps. First, we build a model that predicts the treatment variable based on the input variables . Then, we create a separate model that predicts the outcome variable using the same set of input variables . Subsequently, we calculate the residuals from the former model and regress them against the residuals from the latter model. An…

Not sure what point you are trying to make here. Double ML is a valid approach for debiasing confounding effects.

I disagree. It's vulnerable to all sorts of mishaps. You're now having to worry about data leakage between your treatment group AND your target variable. Casual inference without experiment data is all just a mathematical exercise to make a one size fits all approach to identifying relationships. Yes, correlation has weaknesses. But the name "causal inference" is grossly misleading. It's "well if we assume X, Y, and Z then the effect which we have already assumed is causal is probably around this order of magnitude". And hey, maybe that will help you identify cases where a confounding variable is actually the thing that matters. But you're not going to do better than just doing an analysis on the variables and their interactions. You don't have the brainpower to do this at a scale larger than pretty much all causal methods will begin to fail. It does not offer you the legitimacy the name implies.

I think it confuses far more than it helps.

Re: Causal inference as a blind spot of data scientists

#82

Post by bangkoksbest »

nyssos wrote: post #12
tomrod wrote: post #11

Earlier quoted context omitted.

> Insofar as causal inference has no such 'check', its because there never was any. Casual inference is about dispelling that illusion. Aye, and that's the issue I'm trying to understand. How to know if model 1 or model 2 is more "real" or, for my lack of a better term, more useful and reflective of reality? We can focus on a particular philosophical point, like parsimony / Occam's razor, but as far as I can tell tha…

> How to know if model 1 or model 2 is more "real" or, for lack of a better term, more useful and reflective of reality? You don't. Given observational data alone, it's typically only possible to determine which d-separation equivalence class you're in. Identifying the exact causal structure requires intervening experimentally. > There should be some way to determine a model's likelihood of structure Why? If the info…

Acyclic structure on variables is a very strong pre-supposition that, honestly, is not how many systems in engineering are well-described by, so I don't like this idea of boiling causality solely down to DAG-dependent phrases like "d-separation" or "exact causal structure". Exact causal structure a.k.a. actual causality is particular to one experimental run on one intervention.

Re: Causal inference as a blind spot of data scientists

#83

Post by rubslopes »

g42gregory wrote: post #22

In Corporate and Medical data science fields, people begin to accept causal inference. It is difficult, as the subject is still in flux and under development. I am aware of three reputable causal inference frameworks: 1. Judea Pearl's framework, which dominates in CS and AI circles 2. Neyman-Rubin causal model: https://en.wikipedia.org/wiki/Rubin_causal_model 3. Structural equation modelling: https://en.wikipedia.org…

Economist here. Causal inference is more alive than never, in Economics at least. A publication in an applied top journal practically has to use causal methods.

The DID literature, for instance, has been expanding at the speed of light -- it has never been so hard to keep up as it is now.

Re: Causal inference as a blind spot of data scientists

#84

Post by FreakLegion »

jldugger wrote: post #69
shahbazac wrote: post #48

I’ve tried to understand causal inference several times and failed. Tutorials seem unnecessarily long winded. I wish authors would give simple, to the point examples. Say I have a simple table of outdoor temperatures and ice cream sales. What can the machinery of causal inference do for me in this situation? If it doesn’t apply here, what do I need to add to my dataset to make it appropriate for causal inference? Mor…

> Say I have a simple table of outdoor temperatures and ice cream sales. What can the machinery of causal inference do for me in this situation? Not much. Causal inference works over networks of variables, specifically a DAG. But usually you know more than one variable association, so this is more an issue of pedagogy than the tool itself. Probably the shortest, most persuasive example I can give you is a logical res…

When looking at admissions rates across the entire university, women are less likely to be accepted than men. But when (in this example) you break that down into departments, every department favors women over men.

If every department favored women then the entire university would also favor women. Parity is guaranteed in that scenario. What happened in the Berkeley case is that not every department favored women, and women applied disproportionately to the departments with lower admissions rates (including some that didn't favor them), while men did the opposite.

Re: Causal inference as a blind spot of data scientists

#85

Post by nyssos »

bangkoksbest wrote: post #82
nyssos wrote: post #12

Earlier quoted context omitted.

> How to know if model 1 or model 2 is more "real" or, for lack of a better term, more useful and reflective of reality? You don't. Given observational data alone, it's typically only possible to determine which d-separation equivalence class you're in. Identifying the exact causal structure requires intervening experimentally. > There should be some way to determine a model's likelihood of structure Why? If the info…

Acyclic structure on variables is a very strong pre-supposition that, honestly, is not how many systems in engineering are well-described by, so I don't like this idea of boiling causality solely down to DAG-dependent phrases like "d-separation" or "exact causal structure". Exact causal structure a.k.a. actual causality is particular to one experimental run on one intervention.

D-separation still works for cyclic graphs, it just can't rule out causal relationships between variables that lie on the same cycle. And neither can any other functional-form-agnostic method, because in general feedback loops really do couple everything to everything else.

More rigorously: given a graph G for a structural equation model S, construct a DAG G' as follows

- Find a minimal subgraph C_i transitively closed under cycle membership (so a cycle, all the cycles it intersects, all the cycles they intersect, and so on)

- Replace each C_i with a complete graph C'_i on the same number of vertices, preserving outgoing edges.

- Add edges from the parents of any vertices in C_i (if not in C_i themselves) to all vertices in C'_i

- Repeat until acyclic

d-separation in G' then entails independence in S given reasonable smoothness assumptions I don't remember the details of off the top of my head.

Re: Causal inference as a blind spot of data scientists

#86

Post by bertil »

shahbazac wrote: post #48

I’ve tried to understand causal inference several times and failed. Tutorials seem unnecessarily long winded. I wish authors would give simple, to the point examples. Say I have a simple table of outdoor temperatures and ice cream sales. What can the machinery of causal inference do for me in this situation? If it doesn’t apply here, what do I need to add to my dataset to make it appropriate for causal inference? Mor…

> Or do I need to provide that graph to this function?

You need to do that, and the math can help you measure how much each arrow contributes. The idea that you need to provide your model of the world is strangely not a key part of most introductions, but it’s crucial.

> outdoor temperatures and ice cream sales

That’s too simple: a simple regression can handle that. Causal inference can handle cases with three variables, assuming you provide an interaction graph. Say: your ice cream truck goes either to a fancy neighborhood or a working-class plaza. After observing the weather, you decide where to go, so know that wealth and weather influence sales, but sales can’t influence the other two. Assuming you have data all for cases (sunny/poor, sunny/rich, rainy/poor, rainy/rich), then you can separate the two effects.

Re: Causal inference as a blind spot of data scientists

#87

Post by godelski »

tomrod wrote: post #29
godelski wrote: post #27

Earlier quoted context omitted.

> How to know if model 1 or model 2 is more "real" or, for my lack of a better term, more useful and reflective of reality? I just want to second MJ's points here. You have to remember that 1) all models are wrong and 2) it's models all the way down. Your data is a model: it models the real world distribution, what we might call the target distribution, which is likely intractable and often very different from your d…

I love this comment to bits. Thanks, from a fellow applied researcher embedded in the tech world.

I always love to find/know there are others in the tech world that care about the nuance around evaluation math and not just benchmarks. Often it feels like I'm alone. So thank you!

Re: Causal inference as a blind spot of data scientists

#88

Post by jldugger »

FreakLegion wrote: post #84
jldugger wrote: post #69

Earlier quoted context omitted.

> Say I have a simple table of outdoor temperatures and ice cream sales. What can the machinery of causal inference do for me in this situation? Not much. Causal inference works over networks of variables, specifically a DAG. But usually you know more than one variable association, so this is more an issue of pedagogy than the tool itself. Probably the shortest, most persuasive example I can give you is a logical res…

When looking at admissions rates across the entire university, women are less likely to be accepted than men. But when (in this example) you break that down into departments, every department favors women over men. If every department favored women then the entire university would also favor women. Parity is guaranteed in that scenario. What happened in the Berkeley case is that not every department favored women, an…

Yes, apologies, what I meant by "favored" was that in every department, women applicants were more likely to get an admission than men. But I'm pretty sure the admission rate can still be lower for women overall than men overall, using exactly the same scenario you described. If the sociology department admits 10 percent of applicants and the physics department admits 90, it seems very easy for gender bias in applications to shift women towards 10 and men towards 90, even if the rate is a few percent higher for women.

Re: Causal inference as a blind spot of data scientists

#89

Post by jrumbut »

lr1970 wrote: post #47
mikpanko wrote: post #21

Earlier quoted context omitted.

A/b experiments are definitely a gold standard as they provide true causality measurement (if implemented correctly). However, they are often expensive to run: need to implement the feature in question (which is less than 50% going to work) and then collect data for 1-4 weeks before being able to make the decision. As a result only a small number of business decisions today rely on a/b tests. Observational causal inf…

There can be a real ethical dilemma when applying A/B testing in medical setting. Placing someone with an incurable disease in a control group is condemning them to death while in treatment group they might have a chance. On the other hand, without a proper A/B testing methodology the drug efficacy cannot be established. So far no perfect solution to the dilemma has been found.

> in a control group

The control group gets the current standard treatment, not nothing (in case that was a source of confusion). Plus they typically don't have to pay for it which is a benefit for them.

Large trials today will typically conduct interim analyses and will have pre-defined guidelines for when to stop the trial because the new treatment is either clearly providing a benefit or is clearly futile.

Here is an example of such a study: https://www.ahajournals.org/doi/10.1161/CIRCHEARTFAILURE.111...

Post reply on HN