Neil G
Reputation
867
Top tag
Next privilege 1,000 Rep.
Create new tags
 Jun 28 comment Sufficient statistic for product of $\Gamma(x + a_i)$ If this question interests you, please help me tag it. Jun 28 asked Sufficient statistic for product of $\Gamma(x + a_i)$ Jun 17 awarded Quorum Dec 29 revised Convolution of probability densities. Having trouble integrating over vector added 218 characters in body Dec 28 accepted Convolution of probability densities. Having trouble integrating over vector Dec 28 comment Convolution of probability densities. Having trouble integrating over vector thanks! B is the multinomial beta function, which I think normalizes the distribution correctly. the domain of integration is such that the arguments passed to f are always negative (l - x <= l - l = 0). Dec 26 comment How do you get a Percentage of how far the value is from expected Value FYI the mean is a synonym for expected value Dec 25 awarded Teacher Dec 25 answered Making Change for a Dollar (and other number partitioning problems) Dec 25 revised Convolution of probability densities. Having trouble integrating over vector added 301 characters in body; added 36 characters in body Dec 25 asked Convolution of probability densities. Having trouble integrating over vector Dec 17 accepted What is the $d\sigma/\sigma$ rule? Dec 17 comment What is the $d\sigma/\sigma$ rule? Thank you very much for this answer. I guess I am a 'true believer.' Dec 15 asked What is the $d\sigma/\sigma$ rule? Dec 7 comment Big $\mathcal{O}$ Notation question while estimating $\sum \frac{\log n}{n}$ Can someone repair the title: "Big O notation." Dec 1 comment $e$ to 50 billion decimal places @Arturo, I think I understand, thanks. Dec 1 comment $e$ to 50 billion decimal places @Arturo, My point is that the parallelization and vectorizing problem is essentially solved by the mapreduce algorithm. All one needs is a program that calculates some digits, whether in hex or in decimal. Dec 1 comment $e$ to 50 billion decimal places @J.M. I really don't see that. It's a standard map-reduce with key=decimal positions, value=decimal values. The map-reduce framework does the consolidation automatically. Dec 1 comment $e$ to 50 billion decimal places @J.M. Good point, but calculation of the digits of pi and e are easily parallelizable since the calculation of later digits don't depend on earlier ones. Dec 1 comment $e$ to 50 billion decimal places Nice answer, but why do you say that we are not good at parallel computing? Most of Google's services are proof of excellent parallelization.