I read about this, $$ p(R|r=0,NI_{1}) = {N\choose n+1}{N-R\choose n} \; (6.44)$$ but the probability can not be greater than 1, this expression is an apparent error. $I_{1}$ means the state of knowledge about the urn is that 0 < R < N, while the general sum $S = \sum_{R=0}^N{{R\choose0}{N-R\choose n-r}}= {N+1\choose n+1} $ is given by $I_{0}$ which means we do not know the state of the knowledge about urn at all, so the $S'= S - {N\choose n}\delta(r,0) - {N\choose n}\delta(r,n)$, and the S' should be the denominator. Finally i get the result is $$ p(R|r=0,NI_1) = \frac{N-R\choose n}{N\choose n +1}$$

  • $\begingroup$ This question is not readable without getting back to the book and should be made self-contained. $\endgroup$
    – Xi'an
    Nov 4, 2017 at 14:01
  • $\begingroup$ Yes, when I come across problem while reading the book, i will post a question here. I will make question more clear next time. $\endgroup$
    – hello_god
    Nov 4, 2017 at 14:28

1 Answer 1


You will find this error discussed (along with many others) at the website http://ksvanhorn.com/bayes/index.html. For more information on Jaynes's book, see http://www.etjaynesinfo.com

  • 1
    $\begingroup$ Thank you! It would be better if you would edit this post to cite the specific error that was made, rather than requiring interested readers to follow the links. $\endgroup$
    – whuber
    Oct 31, 2017 at 21:25

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.