fit_poisson_model
is called from intensity_pspline
and performs the iterative algorithm to estimate the model parameters and the
smoothing parameters \(rho\) in the penalized Poisson model.
fit_poisson_model(data, Z, K, ind, verbose = FALSE, control = list())
Model fit.
The binned data.
The (sparse) model matrix where the number of columns must
correspond to the length of the vector of model coefficients theta
.
A (sparse) square penalty matrix of with the same dimension as
theta
.
A list which contains the indices belonging to each smooth term and the linear terms.
If TRUE
, prints information on the process of the fitting
algorithm.
A list of optional arguments which control the convergence of the fitting algorithm. See "Details".
Marc Schneble marc.schneble@stat.uni-muenchen.de
Smoothing parameters are estimated using the generalized Fellner-Schall method (Wood and Fasiolo, 2017).
Wood, S. N. and Fasiolo, M. (2017). A generalized Fellner-Schall method for smoothing parameter optimization with application to Tweedie location, scale and shape models. Biometrics 73 1071-1081.