Maximum likelihood
Also known as: MLE, Maximum likelihood estimation, Maximum likelihood estimator
Maximum likelihood estimation (MLE) chooses $\hat\theta$ to maximise the likelihood $L(\theta;x)=\prod_i f(x_i;\theta)$ for fixed data $x$. Prefer the log-likelihood $\ell=\log L$; solve $\partial\ell/\partial\theta_j=0$ and check a maximum.
Examples: Bernoulli $\hat p=\bar X$; Normal $\hat\mu=\bar x$, $\hat\sigma^2_{\mathrm{MLE}}=n^{-1}\sum(x_i-\bar x)^2$ (biased for $\sigma^2$). Under regularity, MLEs are typically consistent and asymptotically normal. For Normal linear models with Normal errors, MLE coincides with OLS.