When I was working on correlation between x and è in Baysian Method an interesting subject
Took place .
Let’s X has ber(p) and P has Uniform(0,1).
The corr. Ceoff. X and P will be sqrt(3)/3 .and if we assume that
The parameter dist. Unif(1/4,3/4) then corr. Ceof . between X and P
Will decrease to sqrt(3)/6 .i.e when we increase (decrease) parameter space
Then corr. Ceoff.between X and P increase (decrease).
I want to go on this subject. For example what will be happen
If we have more than one observation and its relationship between
Some loss function and Baysian Estimator.
Do you feel this subject is worth –while ?
Please inform me if you have knowledge about this.
Tahanks :amin
---------------------------------
Do You Yahoo!?
Yahoo! Auctions Great stuff seeking new owners! Bid now!
|