3 ms·

For instance?

• DannyBee 14y ago
  So, there are some that require offline calculations (I can't really get into more than this for various reasons).
  • victorh 14y ago
    Not even a Wikipedia link? This was so anti-climactic.
    • DannyBee 14y ago
      Wikipedia does not always have the answers :) Sorry, I fully intended to publish a description, but it turns out at least two of them are the subject of pending patent apps that haven't been published yet, so I can't. In any case, do note there are a wide variety of wilson variants or alternatives that may be applicable to various situations (agresti-coull, etc), as well as jeffrey's prior bayesian intervals. http://stat.wharton.upenn.edu/~tcai/paper/Binomial-Annals.pdf http://stat.wharton.upenn.edu/~tcai/paper/Binomial-Annals.pd... is a reasonable comparison of the current common methods. Anyway, I deleted the original comment since it doesn't add anything now.
  • knite 14y ago
    Perhaps I'm a bit dense today, but why are you unable to link to descriptions of one or more of these better approaches? Many people would find the information useful and interesting.