PDA

View Full Version : normal distibution standard deviation


mittman84
11-14-2005, 04:47 PM
I feel really dumb asking this question, but it is in my assignment that is due tomorrow and i am drawing a blank, and i have to figure out all this stuff by myself since my teacher is worthless. In in a certain normal distibution of X, the probability is .5 that X is less than 500, and the probability is .0227 that X is greater than 650, what is the standard deviation of X?

Indiana
11-14-2005, 04:50 PM
Recheck ur post. U left off the number after the decimal..

Indy

mittman84
11-14-2005, 04:53 PM
thank you, should be good now

edtost
11-14-2005, 06:09 PM
well, you have:
http://mathworld.wolfram.com/images/equations/NormalDistribution/equation1.gif

for two observations of x, and know all parameters except for mu and sigma, which are the same for both observations. plug in numbers, solve for mu, set the expressions for mu equal to each other, solve for sigma.

edit: that's the pdf, so you actually have the integral of that for two observations.

edit edit: one of your observations is the mean. this makes the problem much easier.

mittman84
11-14-2005, 06:49 PM
i came up with 75...does that seem right?

edtost
11-14-2005, 07:20 PM
yup.

based on what the answer turned out to be, can you think of an easier way to do the original problem than doing out all the calculations?

mittman84
11-14-2005, 07:32 PM
yes, z=(X-mean)/sd used the z table to find z=2.0 at .9773
so 650-500/sd = 2 solve and sd =75 that means 650 is two standard deviations from the mean, and because it is normal that will be 95% of the values. It is not exactly 75 because if rounding errors when using the z-table but 1-2*.0227=.95(roughly)tells that 650 is roughly 2 standard deviations from the mean, and mean =500