s.petersson <[EMAIL PROTECTED]> wrote:
> I sometimes run into a constant of 1.96 stdv that is used to calculate 95%
> statistical confidence intervals. But I can't seem to find how the 1.96 stdv
> is actually derived from a security level of 95%. In the statistical
> textbooks I've read, there is only a huge table with different stdv's at a
> given security level.
It comes from the fact that 95% of the area under a standard normal curve
falls into the interval (-1.96,1.96).
> Let's say I want to calculate this constant with a security level of
> 93.4563, how do I do that? Basically I want to "unfold" a function like
> this:
> f(95)=1.96
> Where I can replace "95" with any number ranging from 0-100.
You consult a standard normal table to find the interval that contains x%
of the area under the curve. Since most standard normal tables start at
zero due to the symmetry of the distribution, you first need to "correct"
x by dividing it by 200 and adding .5.
So, for x=95, you transform it to x/200+.5=.975 and then you look at the
table to see what value of z has a cumulative probability of .975. The
closest z-value is 1.96. For 93.4563, you look up the z-value for .96729,
which is about 1.84.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
