Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If I multiply $N$ irrational numbers together to generate a product $P$, where the irrational number are specified to a working precision of $m$ decimal digits, how many decimal digits, $k$, should I have confidence in for $P$?

For example, how large of an $N$ can I get away with standard double floating point precision to have a confidence of $k$ decimal digits in the product (about m ~ 15 stable decimal places)?

share|cite|improve this question

Roughly speaking, if you have an error of a factor of $1 \pm 10^{-15}$ in each multiply, the standard deviation is a factor of $1 \pm \frac 1{\sqrt 3}10^{-15}$ (assuming a uniform distribution). You need about 3 of these to lose the first digit, 300 to lose the second.

It isn't worth being too careful in this calculation, because if your number is stored in decimal, the fractional error changes by a factor 10 between a number that starts 10000 and 99999. In binary it is only a factor of 2, but the principle is the same.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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