15

TL;DR

I solved two integrals left unsolved in a paper I'm reading. Should I notify the author?


I am studying in a finance master and we have been assigned a project for one of our class. This project consisted in some literature review and then a small implementation.

Reading one of the papers, in which the author propose a new density p to model a random variable, there is written that

  • They can't find analytically the constant c that makes p a density (basically could not compute the integral of p)
  • They write that numerical simulations show that E[X]= theta but an analytical proof was still missing

I have calculated in closed form both integrals, and numerical simulations suggest that my results are correct; I will probably add this to the project I'm writing.

Now the question is: should I send an email to the author about this? If so, how to phrase such an interaction? I guess that I'm looking for something along the lines of "thank you, well done" with maybe a short mention if they get around to modify the original paper.

On the other hand, the paper is old (2007) so I don't even know if the author is still interested in that.

Also, the proof was nothing too complicated. I mean I spent a lot of time on it (mainly because of my inexperience) but it certainly doesn't feel like anything too difficult. Probably I just "saw" the right path to do it and once you see that, it's pretty easy.

Is this something common to do? Are professors pleased to receive such an email or just annoyed?

  • On top of what has been said, even if the paper is 8 years old, the professors are quite possibly still working on related questions, using similar methods. I could easily imagine cases where this would be great news to the authors as it helps making current research a bit more complete, or at least easier. As mentioned elsewhere, you might be acknowledged for your help then. In any case, I see no harm in contacting them. – gnometorule Nov 12 '15 at 17:21
9

The significance of your additional work is unclear. Have you talked about this with the professor that teaches your class? (s)he – or another prof in the department – should be involved if you want to take this further.

If the additional work is significant enough, you'll be perfectly within your rights to write and publish your own paper that presents the results. Trivial results usually aren't publishable on their own, but if it's a novel approach then it's certainly doable. The original work is old enough that you need not worry about "scooping" the original authors' results. Just make sure to check that a solution to the problem hasn't already been published elsewhere.

In any other case, there shouldn't be any harm in sending an email to the authors (whether or not they will respond is another matter). Don't send the solution in your first email! Just inform the authors that you think that you've solved the problem, and only send them your solution if they agree to let you co-author any publications that incorporate your results. Intellectual property rights are a tricky business, and you should always ask for the help of a professor when doing things like this. The paper's authors will also be much more likely to take notice if the first email is from a well known professor, rather than an unknown grad student.

  • first of all, thank you very much for the answer. I certainly don't think that it's publishable on his own, and I don't even think it is worth of co-autorship with the original paper. I have never published anything but it would be cool to have an acknowledgement , like "we thank Ant for showing that " or something like that ? – Ant Nov 12 '15 at 14:38
  • 4
    This is generally good advice, but I think it is definitely important to determine the significance. If the work is pretty trivial, even though it may be interesting to the other authors, it will sound silly if you go into this discussion of co-authorship. And insisting on co-authorship on any publication that incorporates the result is extreme: the result itself may not be enough of a contribution to warrant authorship, and it's not clear whether the OP would be qualified to contribute in other ways. In many cases a simple acknowledgement would be fine. – Nate Eldredge Nov 12 '15 at 15:11
  • 1
    @NateEldredge thinking the same thing at the same time! The OP really should discuss this with a professor. Which actions are appropriate really depends upon how significant the solutions are. – Moriarty Nov 12 '15 at 15:13
  • 2
    @NateEldredge Thank you both. So I really need to check with the professor; I'll ask him. In any case I think it's (probably) trivial for professor or real mathematicians, that is why it struck me weird at first that the author did not find it (elsewhere on the article there are very complex derivations of other stuff which I wouldn't be able to prove, for example). On the other hand I am quite certain that my proof is correct.. Oh well. I'll just ask the professor :-) – Ant Nov 12 '15 at 16:51
  • 4
    @Ant Interesting question Ant. Could you keep us updated on what transpires? – user42055 Nov 13 '15 at 3:20

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.