Most Likelihood Estimation

1. Likelihood

a function of the parameters of a statistical model given data

- Discrete probability distribution

${\displaystyle {\mathcal {L}}(\theta \mid x)=p_{\theta }(x)=P_{\theta }(X=x)}$

- Continuous probability distribution

${\displaystyle {\mathcal {L}}(\theta \mid x)=f_{\theta }(x)}$

2. MLE

a method of estimating the parameters of a statistical model, given observations

The method defines a maximum likelihood estimate:

$\hat\theta_{\text{MLE}} = \arg \max_{\theta} L(\theta; {x_i})$

natural logarithm of the likelihood function, called the log-likelihood:

or the average log-likelihood:


Reference

공돌이의 수학정리 노트

통계강의록 1


© 2018. All rights reserved.

Powered by Hydejack v7.5.0