Definition: Maximum log-likelihood
Index:
The Book of Statistical Proofs ▷
General Theorems ▷
Frequentist statistics ▷
Likelihood theory ▷
Maximum log-likelihood
Sources:
Metadata: ID: D61 | shortcut: mll | author: JoramSoch | date: 2020-05-15, 23:13.
Definition: Let there be a generative model $m$ describing measured data $y$ using model parameters $\theta$. Then, the maximum log-likelihood (MLL) of $m$ is the maximal value of the log-likelihood function of this model:
\[\label{eq:mll} \mathrm{MLL}(m) = \operatorname*{max}_\theta \mathrm{LL}_m(\theta) \; .\]The maximum log-likelihood can be obtained by plugging the maximum likelihood estimates into the log-likelihood function.
Metadata: ID: D61 | shortcut: mll | author: JoramSoch | date: 2020-05-15, 23:13.