What would be minimal test-case(s) and corresponding solution, demonstrating curent know-how in causation/precedence analysis solving, or more simply what is it possible to achieve using causation/precedence analysis techniques since the field seems to have evolved fast during the last decade:

Statistics calculations have a cost and are often used to convince rather than used as a way to know the truth. May-be is it because actual hability of statistics to analyse facts is still limited to correlation rather than causation. SEM path analysis, went popular surprizingly in social, behavioural science and psychology, trough the use of LISREL, but seems still not be used to build survey or analyse survey results. later, Judea Pearl in 2011 and Leslie Lamport 2013 won the ACM Turing award (IT Nobel) on subjects related to causation.

Artificial intelligence needs to optimize some form of truth, using causation. But its optimisation process should be improved by including intuitive-like strategies where comfirmatory hypothese analysis would help a lot, ..."I believe". I imagine that Causation analysis is part of recent IA achievments in a way or another. But practically how ? What test-case were solved ?

  • 3
    $\begingroup$ I'm voting to close this question as off-topic because this question is about statistics and would better fit on Cross Validated. $\endgroup$ – Robin Kramer Apr 28 '17 at 8:00
  • 1
    $\begingroup$ updated the content to explain the implication with cognitive sciences $\endgroup$ – user1767316 Apr 28 '17 at 12:26
  • $\begingroup$ How is "state of the art" defined here? I don't mean to be facetious, I genuinely don't know exactly what exactly you'd be trying to demonstrate. $\endgroup$ – ancientcampus Apr 28 '17 at 13:47
  • $\begingroup$ Agree with Robin Kramer. This is off-topic here. $\endgroup$ – Spero Apr 28 '17 at 16:30
  • $\begingroup$ @user1767316 I think the question may be very interesting, but I am afraid you will not receive a satisfying answer here. By migrating it to Cross Validated the question will not disappear from this site, but an entire new audience (that may have more experience with LISREL and multivariate analyses) will be able to look at the question and formulate an answer $\endgroup$ – Robin Kramer Apr 29 '17 at 12:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.