5
$\begingroup$

This seems like an exciting approach to uplift modelling, but the only resource that I can find is this paper and it is too brief, notation-heavy and dense to be of any use to me.

I have an honours degree in mathematical statistics, but it is not nearly enough to decipher this paper. Irefuse to put methodologies into production that I can not on any level understand.

Are there people who could code this algorithm themselves based on the material presented in this paper? Can somebody give an intuitive/simpler description?

Improve this question
$\endgroup$
6
  • 3
    $\begingroup$ I agree this paper is not an easy read, and also that it is very interesting. Not to put you off, but in lots of cases, the people putting these kinds of papers into practice will have quite a few years of postgraduate education, at masters level, doctorate and beyond. If there are any specific parts of the paper you are having problems with you might get good answers on here or other SE sites, but I'm sorry to say that as it stands this question might get closed for needing clarity. $\endgroup$
    – Robert Long
    Commented Sep 26, 2020 at 19:01
  • 4
    $\begingroup$ I edited the format and positioning of your question. I find it an interesting question. But maybe the question 'does anybody understand this article?' is not so clear. I suggest that you edit a bit further adding your current understanding and maybe pinpoint a specific point where it gets confusing. $\endgroup$
    – Sextus Empiricus
    Commented Sep 26, 2020 at 19:41
  • 3
    $\begingroup$ Is this the article? arxiv.org/abs/1610.01271 $\endgroup$
    – Sextus Empiricus
    Commented Sep 26, 2020 at 19:42
  • 2
    $\begingroup$ +1 for refusing to put methodologies into production that you don't understand understand. (You might want to see these two threads too: stats.stackexchange.com/questions/370033 & stats.stackexchange.com/questions/420529) $\endgroup$
    – usεr11852
    Commented Sep 26, 2020 at 20:21
  • 1
    $\begingroup$ I suppose you might get some clues by reading the code in the package on CRAN. Sometimes trying to understand them side by side helps. Sometimes not of course. I think you may have to break this down into sections and ask about each part you do not understand (as @SextusEmpiricus suggests). $\endgroup$
    – mdewey
    Commented Sep 27, 2020 at 14:23

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.