To construct the half-normal probability plot, plot the absolute values in a certain statistical diagnostic (residual, leverage, Cook distance and others) versus $z_i$ where:

$\displaystyle z_{i} = \Phi^{-1} \left(\frac{k + n - \frac{1}{8}}{2n +\frac{1}{2}}\right)$

$\therefore$ Applied Linear Statistical Models, (Neter, 1996) Fifth Edition, Pag 596


$\displaystyle z_{i} = \Phi^{-1} \left(\frac{k + n - \frac{1}{2}}{2n +\frac{9}{8}}\right)$

$\therefore$ Generalized Linear Models, (Nelder, 1989) Second Edition, Pag 407

where $\Phi$ is the cumulative distribution function of the standard normal distribution.

$k$ is the $kth$ ordered absolute residual

$n$ is the sample size

I could not locate the mathematical justification for the difference of the $z_i$, and more importantly I don't have any intuition about where it comes from.

Any help? Could you please guide me?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.