loglik = new(loglik_gda, om, terms, y, x)
This is a standard model which has the form
$$y = \langle \phi(x), \theta \rangle + \delta(x) + \varepsilon,
\delta(x) \sim N(0, \lambda g(x)), \varepsilon \sim N(0,\sigma^2)$$
where \(\phi(x)\) is the basis, \(\theta\) is the coefficient vector,
\(\delta(x)\) is unseen vector corresponding to unmodeled
variance \(\lambda g(x)\), \(\varepsilon\) is an unseen noise vector.
The parameter vector is of length 2 where
\(\sigma=\) exp(para[0]) and \(\lambda=\)exp(2*para[1]).
an outermod instance to be referred to
a matrix of terms, must have as many columns as dims in
om
a vector of observations
a matrix of predictors, must have as many columns as dims in
om and the same number of rows as y
no returns, this is a class which contains methods
base class: lpdf