What is the generalized R-Squared?

Nov 1, 2015

${R}_{\text{GEN}}^{2} = 1 - {\left(\frac{L \left(0\right)}{L \left(\hat{\theta}\right)}\right)}^{\frac{2}{n}}$

Explanation:

The generalised ${R}^{2}$ is defined as

${R}_{\text{GEN}}^{2} = 1 - {\left(\frac{L \left(0\right)}{L \left(\hat{\theta}\right)}\right)}^{\frac{2}{n}}$

where $L \left(0\right)$ : the likelihood of the model with only the intercept

$L \left(\hat{\theta}\right)$ : the likelihood of the estimated model
and $n$ : sample size.

Also See this