P(X[M+1}|X[1],...,X[M])
= integral from 0 to 1 of P(X[M+1]|theta)P(theta|X[1],...,X[M])dtheta
= integral from 0 to 1 of
P(X[M+1]|theta)P(X[1],...,X[M]|theta)P(theta)dtheta/P(X[1],...,X[M])
= integral from 0 to 1 of theta(theta)sup4(1-theta)1dtheta/30
= 5/7
Of course, there is a theorem where you do not the need to go back to
basics like I just did. The notation is horrible using this email word
processor. If you like, I can send you a postscript file with the same.
Rich
At 01:50 PM 12/29/98 +0800, research2@usa.net wrote:
>I refer to the updated tutorial "Learning Bayesian Networks from Data"
>by Nir & Moises. Website: http://www.cs.berkeley.edu/~nir/Tutorial/
>
>In slide 47 titled "Example: Binomial Data Revisited", I would like to
>know how the value 5/7 was calculated based on the integration
>formula.
>
>Thanks.
>
>