Something's value changes randomly over time. Assume normal (Gaussian) distribution. I know that it spends X% of its time greater than BIGNUMBER and (equally) X% of its time smaller than SMALLNUMBER. It's symetric, not skewed or anything weird like that, and so the mean is the mean of BIGNUMBER and SMALLNUMBER.
What's the standard deviation??
Thanks!!!
2006-09-14 23:07:48 · 8 answers · asked by wild_eep 6 in Science & Mathematics ➔ Mathematics
You can find the std deviation (sigma) by solving for it in this equation
1/2 - X/100 = erf((BIGNUM-mean) / sqrt(2) / sigma)
It has X/100 if X is in %. erf is the error function. You can solve for sigma using a numerical method.
2006-09-15 00:58:27 · answer #1 · answered by Seryt 2 · 1⤊ 0⤋
The standard deviation depends on what x% was.
For example, in a normally distributed function the proportion of data within one standard deviation of the mean is 68 percent and within two standard deviations is 95 percent and so on.
So if x =68 then the SD is BIG NUMBER minus the mean you worked out.
If x=95 then the the SD is (BIG NUMBER - mean)/2
I hope that is helpful :)
2006-09-15 07:58:24 · answer #2 · answered by Tammi J 3 · 1⤊ 0⤋
Use your normal distribution tables and work backwards, as you know the area under the curve and the mean, so you can work out the variance and then the sd
It is long time since I studied stats, all of my notes and texts books are at my parents house.
2006-09-15 06:22:12 · answer #3 · answered by Bob 3 · 1⤊ 0⤋
Who cares!
2006-09-15 06:16:31 · answer #4 · answered by Recko 2 · 0⤊ 2⤋
Promise me you will get some help soon !
2006-09-15 06:19:21 · answer #5 · answered by Robert B 3 · 0⤊ 2⤋
you what speak english man
2006-09-15 06:11:08 · answer #6 · answered by locko550 2 · 0⤊ 2⤋
sorry, am not that good in maths....
2006-09-15 06:15:25 · answer #7 · answered by Anonymous · 0⤊ 2⤋
what?
2006-09-15 06:15:57 · answer #8 · answered by Anonymous · 0⤊ 2⤋