David,
I have the fifth edition, but I am sure things are done the same way in both editions. There is no contradiction; however Hogg and Graig define the sample variance as sum(x - xbar)^2 / n as opposed to dividing by (n-1) as most others do. by doing this, Hogg and Craig need to use sqrt(n-1) in their formula while others use sqrt(n). It all balances in the end.
Howard Kaplon
-----Original Message-----
From: David Heiser [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, April 30, 2002 4:41 PM
To: [EMAIL PROTECTED]
Subject: Is Hogg and Craig Wrong?
Been doing a little cross checking between sources on the confidence
interval about sample means, where the sample (n) comes from a normal
population with unknown mean and unknown variance.
Several textbooks give the confidence interval about the sample mean as
t*s/sqrt(n), where n is the number in the sample, t is the t distribution
value at n-1 df, and s is the sample unbiased standard deviation. Excel also
calculates the confidence interval about the mean based on this equation.
Fisher says that s/sqrt(n) is t distributed, and on page 122 (Statistical
Methods) calculates the interval as t*s/sqrt(n).
Hogg and Craig (4th edition) on page 214 (and text material previous to
this, page 144) gives the interval as t*s/sqrt(n-1). Hogg and Craig on page
145 says that the t distribution parameter is degrees of freedom (n-1). I
know this is "small potatoes", but can anybody explain the differences
between sources on the interval size?
DAHeiser
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
