Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

I know this is rather simple, but despite my best efforts and quite a bit of searching through google I can't seem to find a satisfactory answer.

An example of what I'm trying to do:

Given a value $N= 6.8\cdot10^{-4}$ and upper and lower errors for that value of $N^{uperr} = 9.1\cdot 10 ^{-4}$ and $N_{lowerr} = 3.9\cdot 10 ^{-4}$

With an equation such as $1.6\cdot 10^{-9}Nx$

How would I go about plotting this point at $x=1$ on a log-log scale, with error-bars given by those I've specified?

Whenever I try to plot this I get error bars which are far too large, since from what I've read online simply adding $N+N^{uperr}$ doesn't work for log-log plots.

share|cite|improve this question

I'll only note that when performing an arbitrary transformation $f(u)$ on an uncertain number of the form $x\pm\epsilon$, the formula you'll need goes like

$f(x\pm\epsilon)=f(x)\pm f^{\prime}(x)\epsilon$

(hint: Taylor)

Now apply this to the transformations needed for doing log-log plots...

share|cite|improve this answer
Thanks, I've actually tried this formula on my data before, but i was still getting error bars that were much larger than i was expecting. After trying it on some other plots I'm fairly certain the problem was more with the plot I was trying to emulate than with the error bars. – user7023 Feb 13 '11 at 14:18

Your Answer


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