summary.lgspline {lgspline} | R Documentation |
Summary method for lgspline Objects
Description
Summary method for lgspline Objects
Usage
## S3 method for class 'lgspline'
summary(object, ...)
Arguments
object |
An lgspline model object |
... |
Not used. |
Value
An object of class summary.lgspline
. This object is a list
containing detailed information from lgspline
fit, prepared for
display. Its main components are:
- model_family
The
family
object or custom list specifying the distribution and link.- observations
The number of observations (N) used in the fit.
- predictors
The number of original predictor variables (q) supplied.
- knots
The number of partitions (K+1) minus 1.
- basis_functions
The number of basis functions (coefficients) estimated per partition (p).
- estimate_dispersion
A character string ("Yes" or "No") indicating if the dispersion parameter was estimated.
- cv
The critical value (
critical_value
from the fit) used by theprint.summary.lgspline
method for confidence intervals.- coefficients
A matrix summarizing univariate inference results. Columns typically include 'Estimate', 'Std. Error', test statistic ('t value' or 'z value'), 'Pr(>|t|)' or 'Pr(>|z|)', and confidence interval bounds ('CI LB', 'CI UB'). This table is fully populated only if
return_varcovmat=TRUE
was set in the originallgspline
call. Otherwise, it defaults to a single column of estimates.- sigmasq_tilde
The estimated (or fixed) dispersion parameter,
\tilde{\sigma}^2
.- trace_XUGX
The calculated trace term
\text{trace}(\mathbf{XUGX}^T)
, related to effective degrees of freedom.- N
Number of observations (N), re-included for convenience and printing.