Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. 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 have a pulse profile (binned photon counts versus phase) of a star, and for each count rate I have its statistical error.

I want to calculate the so-called pulsed-fraction $P_{frac}=\frac{F_{max}-F_{min}}{F_{max}+F{min}}$, where $F_{max}$ and $F_{min}$ are the maximum and minimum count rates, respectively.

Is it possible to do propagation error to find the error for $P_{frac}$? If yes, how?

share|cite|improve this question
What kind of errors do your measurements have. I assume standard deviation. If not it might influence the method of error propagation. – fibonatic Mar 16 '14 at 11:53
I am sorry, do you mean: what kind of distribution of the photon is assumed? If so, it is gaussian, yes. Otherwise please, could you explain me a little bit better what you mean? – Py-ser Mar 16 '14 at 12:09
up vote 1 down vote accepted

for simplicity of notation say $P = \frac{X - N}{X + N}$

given $\delta X$ is the uncertainty in X and $\delta N$ is the uncertainty in N


$\delta (X - N)$ = $\delta (X + N) = \sqrt {\delta ^2X + \delta ^2N}$

and therefore:

$\delta P = P \sqrt{(\frac{\sqrt {\delta ^2X + \delta ^2N}}{X - N})^2 + (\frac{\sqrt {\delta ^2X + \delta ^2N}}{X + N})^2}$

This is based upon equations 1b and 2b of the following reference:

share|cite|improve this answer
And this is the $1\sigma$ confidence level, right? – Py-ser Mar 17 '14 at 8:26
The propagation formula isn't limited to a particular number of standard deviations. If the uncertainity in Fmax and Fmin represent 1-sigma, then the unceratinty the formula gives will reflect 1-sigma. If the uncertainity is Fmax and Fmin represent 3-sigma, then the unceratinty the formula gives will reflect 3-sigma. You have to be consistent in the way you express the uncerainty for Fmax and Fmin; if you use n-sigma for Fmax, you have to use n-sigma for Fmin if you use the error propagation formula. – DavePhD Mar 17 '14 at 14:08
thanks a lot! One last thing: can the statistical uncertainty I have on my data points be a standard deviation as well? I haven't understood the question of the user here. – Py-ser Mar 18 '14 at 10:33
user fibonatic was asking, when you say "I have the statistical error", is this statistical error the standard deviation. The formula I gave you is based upon your error in Fmax and error in Fmin being a standard deviation or some particular multiple of the standard deviation, like 3 standard deviations. For example see equations 1b and 2b versus equations 1a and 2a in the referecnce I cited. – DavePhD Mar 18 '14 at 11:49

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.