IMDB uses a Bayesian estimate to compute its top 250 films. Let's say I know all the variables needed to compute this Bayesian estimate for these films. I also have the production company associated with each film. How could I fairly rank the production companies based on the weighted rank (and review counts) of their films? I'd be using all films with a more than m reviews. As a refresher, IMDB uses weighted rating (WR) = (v / (v+m)) × R + (m / (v+m)) × C.

It probably doesn't matter, but I'm using R to do the analysis in case there are some packages that can help solve this.


1 Answer 1


Cool question!

What if you created a weighted ranking for each of the production studios using the same formula IMDB use for calculating film rankings? Then you could just sum the components across the films studios. You could think of it as though each studio released a festival showing only its own films, and you were ranking the festivals.

  • $\begingroup$ Of course, this has problems. It depends on the weight you give to a production company that produces lots of fairly good films relative to a company that makes one or two films that rank very highly. $\endgroup$
    – Jim
    Commented Oct 23, 2012 at 3:00
  • $\begingroup$ Thanks. I'm going to wait a bit and see if there are any other answers. I too had this idea, but wasn't familiar enough with the subject to know whether there were any problems with it. $\endgroup$
    – Bradford
    Commented Oct 23, 2012 at 12:44

Your Answer

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

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