Poisson-Gamma Additive Models
AIC extraction
Backfitting algorithm
Smoothing of nonparametric terms
Coefficients extraction
Deviance extraction
Utility function
Normal plot with simulated envelope of the residuals.
Generic function for simulated envelope generation.
Utility function
Fitted values extraction
Utility function
Read the model formula and split it into the parametric and nonparamet...
Utility function
Utility function
Utility function
Loglik extraction
Utility function
Raw Periodogram
Estimation of the conditional distributions parameters of the level
One-step ahead prediction and variance
Utility function
Likelihood function to be maximized
Utility function
Utility function
Poisson-Gamma Additive Models
Plot of estimated curves
Prediction
Model output
Summary output
Residuals extraction
Summary output
LaTeX table exporter
This work is an extension of the state space model for Poisson count data, Poisson-Gamma model, towards a semiparametric specification. Just like the generalized additive models (GAM), cubic splines are used for covariate smoothing. The semiparametric models are fitted by an iterative process that combines maximization of likelihood and backfitting algorithm.