Log-likelihood

Log-Likelihood
The likelihood is defined as a product, and maximizing a product is usually more difficult than maximizing a sum.

The log-likelihood of the null model (L0), that is, the model containing the intercept only (and no regression coefficients) is computed as:
log(L0) = n0*(log(n0/n)) + n1*(log(n1/n)) ...

denotes the log-likelihood of x from a multivariate normal distribution with mean m and any positive definite covariance matrix C.

See also: Likelihood, Distribution, Variance, Estimation, Normal distribution