“…If g denotes time point g = max(p, m, k) + 1, the joint probability density function of the random variables {e g+1 , e g+2 , ... , e N } given by: (7) As the Jacobian of the transformation from {e g+1 , e g+2 , ... , e N } to {x g+1 , x g+2 , ... , x N } is unity, the likelihood function of {x g+1 , x g+2 , ... , x N } is the same as the joint probability density function of {e g+1 , e g+2 , ... , e N }. In light of (7), maximization of the likelihood function is approximately equivalent to minimization of the function Q(⍜) 8) with respect to the parameter vector ⍜. Minimization of Q(⍜) with respect to ⍜ yields the conditional maximum likelihood estimate of ⍜, conditioned on the data {x t }.…”