layout | mathjax | author | affiliation | e_mail | date | title | chapter | section | topic | theorem | sources | proof_id | shortcut | username | |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
proof |
true |
Joram Soch |
BCCN Berlin |
joram.soch@bccn-berlin.de |
2020-01-23 15:38:00 -0800 |
Conjugate prior distribution for binomial observations |
Statistical Models |
Count data |
Binomial observations |
Conjugate prior distribution |
|
P29 |
bin-prior |
JoramSoch |
Theorem: Let
Then, the conjugate prior for the model parameter
Proof: With the probability mass function of the binomial distribution, the likelihood function implied by \eqref{eq:Bin} is given by
In other words, the likelihood function is proportional to a power of
The same is true for a beta distribution over
the probability density function of which
exhibits the same proportionality
and is therefore conjugate relative to the likelihood.