View Full Version : Variance - Course 4
10-26-2001, 01:03 PM
How do you know when to adjust the variance for degrees of freedom?
10-26-2001, 01:08 PM
When you use the sample mean as the M in
Generally, you reduce DF whenever you use an estimated parameter in the estimation of another parameter.
I'm not a PhD in statistics, so see what others say in addition to mine.
<font size=-1>[ This Message was edited by: AOnly on 2001-10-26 13:08 ]</font>
10-26-2001, 01:44 PM
Thanks. That makes sense. In Mahler's Conjugate Priors packet, problem #1.39, he solves the problem by adjusting for the degrees of freedom, but then notes that the original solution for this problem when it was on the exam did not adjust for degrees of freedom...
New at pd
10-26-2001, 01:47 PM
What was the problem?
10-26-2001, 01:53 PM
You are given the following:
The number of losses arising from m+4 individual insureds over a single period of observation is distributed as follows:
Number of Losses Number of Insureds
3 or more 0
The number of losses for each insured follows a Poisson distribution, but the mean of each such distribution may be different for individual insureds.
The variance of the hypothetical means is to be estimated from the above observations.
Determine all values of m for which the estimate of VHM will be greater than 0.
Answer: Adjusting for degrees of freedom yields answer D. Not adjusting is E.
Just realized this is also sample exam question #39.
vBulletin® v3.7.6, Copyright ©2000-2013, Jelsoft Enterprises Ltd.