Fitted bgeva object

Share:

Description

A fitted Binary Generalized Extreme Value Additive object returned by function bgeva and of class.

Value

fit

A list of values and diagnostics extracted from the output of the algorithm. For instance, fit$argument and fit$S.h return the estimated parameters and overall penalty matrix scaled by its smoothing parameters, for the model. See the documentation of trust for diagnostics.

coefficients

The coefficients of the fitted model provided as follows. Parametric and regression spline coefficients.

gam.fit

A univariate logistic additive model object. See the documentation of mgcv for full details.

sp

Estimated smoothing parameters of the smooth components for the fitted model.

fp

If TRUE, then a fully parametric model was fitted.

iter.sp

Number of iterations performed for the smoothing parameter estimation step.

iter.if

Number of iterations performed in the initial step of the algorithm.

iter.inner

Number of iterations performed inside smoothing parameter estimation step.

tau

The tail parameter of the link function.

n

Sample size.

X

It returns the design matrix associated with the linear predictor.

Xr

It returns the design matrix actually used in model fitting.

good

It returns a vector indicating which observations have been discarded in the final iteration.

X.d2

Number of columns of the design matrix. This is used for internal calculations.

l.sp

Number of smooth components.

He

Penalized hessian.

HeSh

Unpenalized hessian.

Vb

Inverse of the penalized hessian. This corresponds to the Bayesian variance-covariance matrix used for ‘confidence’ interval calculations.

F

This is given by Vb*HeSh.

t.edf

Total degrees of freedom of the estimated model. It is calculated as sum(diag(F)).

bs.mgfit

A list of values and diagnostics extracted from magic.

conv.sp

If TRUE then the smoothing parameter selection algorithm converged.

wor.c

It contains the working model quantities given by the square root of the weight matrix times the pseudo-data vector and design matrix, rW.Z and rW.X.

eta

The estimated linear predictor.

logL

It returns the value of the (unpenalized) log-likelihood evaluated at the (penalized or unpenalized) parameter estimates.

Author(s)

Maintainer: Giampiero Marra giampiero.marra@ucl.ac.uk

See Also

bgeva, plot.bgeva, summary.bgeva