Re: adding probabilities
From: packat (chariya_at_verizon.not)
Date: 03/26/05
- Next message: packat: "Re: Naive Bayesian Algorithm?"
- Previous message: Richard Ulrich: "Re: Estimating r^2 from F ratio"
- In reply to: Predictor: "Re: adding probabilities"
- Messages sorted by: [ date ] [ thread ]
Date: Sat, 26 Mar 2005 16:32:48 GMT
Predictor wrote:
> Given no more information than provided in your original
> message, one
> can only infer that the probability of A, by either
> mechanism is "at
> least 0.9".
>
> The other information needed is the relationship between
> mechanisms 1
> and 2 (their "correlation"). It is possible that
> mechanism 2 only
> causes event A when mechanism 1 already has, which would
> result in a
> probability of 0.9.
>
> If it is the case that mechanism 2 generates event A in
> all cases when
> mechnism 1 does not, then the probability is 1.0.
>
> If the outcome of mechanisms 1 and 2 are uncorrelated,
> then the
> formula you derived is correct, although it is usually
> re-arranged
> algebraically as (using your notation):
>
> P1 + P2 - (P1 * P2)
>
In a more general situation, please search Google for
Dempster-Shaeffer theory.
pac
>
> -Will Dwinnell
> http://will.dwinnell.com
- Next message: packat: "Re: Naive Bayesian Algorithm?"
- Previous message: Richard Ulrich: "Re: Estimating r^2 from F ratio"
- In reply to: Predictor: "Re: adding probabilities"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
