Distinguishing Cause From Effect Using Observational Data
1–10 of 10 posts
Re: Distinguishing Cause From Effect Using Observational Data
#2Re: Distinguishing Cause From Effect Using Observational Data
#3https://arxiv.org/abs/1412.3773 https://medium.com/the-physics-arxiv-blog/cause-and-effect-t...
Thanks, we updated the link from http://www.vocativ.com/335705/correlation-causation to this.
Re: Distinguishing Cause From Effect Using Observational Data
#4so p0.8, wasn't 0.95 the standard for claims once?
Re: Distinguishing Cause From Effect Using Observational Data
#5so p0.8, wasn't 0.95 the standard for claims once?
No, because a classification accuracy is not a p-value. By construction, a random guesser would achieve 50% accuracy in guessing whether A~>B or A50% accuracy is the goal here.
Re: Distinguishing Cause From Effect Using Observational Data
#6There is a popular science book on this topic as well:
Why: A Guide to Finding and Using Causes
https://www.amazon.com/Why-Guide-Finding-Using-Causes-ebook/...
It is sitting on my bookshelf, but I haven't managed to get to reading it yet.
Re: Distinguishing Cause From Effect Using Observational Data
#7so p0.8, wasn't 0.95 the standard for claims once?
No, because a classification accuracy is not a p-value. By construction, a random guesser would achieve 50% accuracy in guessing whether A~>B or A 50% accuracy is the goal here.
Interestingly, the authors do acknowledge on p. 46 that their sample size is too small to obtain a statistically significant result:
A rough estimate how large the CauseEffectPairs benchmark should have been in order to obtain significant results can easily be made. Using a standard (conservative) Bon- ferroni correction, taking into account that we compared 37 methods, we would need about 120 (weighted) pairs for an accuracy of 65% to be considered significant (with two-sided testing and 5% significance threshold). This is about four times as much as the current number of 37 (weighted) pairs in the CauseEffectPairs benchmark. Therefore, we sug- gest that at this point, the highest priority regarding future work should be to obtain more validation data, rather than developing additional methods or optimizing computation time of existing methods. We hope that our publication of the CauseEffectPairs benchmark data inspires researchers to collaborate on this important task and we invite everybody to contribute pairs to the CauseEffectPairs benchmark data.
Re: Distinguishing Cause From Effect Using Observational Data
#8Earlier quoted context omitted.
No, because a classification accuracy is not a p-value. By construction, a random guesser would achieve 50% accuracy in guessing whether A~>B or A 50% accuracy is the goal here.
Interestingly, the authors do acknowledge on p. 46 that their sample size is too small to obtain a statistically significant result: A rough estimate how large the CauseEffectPairs benchmark should have been in order to obtain significant results can easily be made. Using a standard (conservative) Bon- ferroni correction, taking into account that we compared 37 methods, we would need about 120 (weighted) pairs for an…
If you want to distinguish a particular method, but you can definitely tell that overall, the methods are collectively outperforming chance and so in this dataset, it is possible to infer the direction of causation.
Re: Distinguishing Cause From Effect Using Observational Data
#9This has been discussed in the stats literature for awhile now. It's an interesting idea but makes lots of assumptions about the nature of noise versus signal. It could be really useful in some situations, but in others it would be totally useless, depending on how realistic the assumptions are in any given scenario.
Re: Distinguishing Cause From Effect Using Observational Data
#10Is the reasoning behind the method a corollary to compressed sensing?