print.smoothSurvReg: Summary and Print for Objects of Class 'smoothSurvReg'

Description

Print a summary information of the fitted model.

For regression coefficients the following information is given:

`Value`	-	estimate of the coefficient
`Std.Error`	-	estimated standard error based on the pseudo-variance estimate (3.1)
		in Komárek, Lesaffre and Hilton (2005)
`Std.Error2`	-	estimated standard error based on the asymptotic variance estimate (3.2)
		in Komárek, Lesaffre and Hilton (2005)
`Z`	-	the Wald statistic obtained as `Value` divided by `Std.Error`
`Z2`	-	the Wald statistic obtained as `Value` divided by `Std.Error2`
`p`	-	the two-sided P-value based on normality of the statistic `Z`
`p2`	-	the two-sided P-value based on normality of the statistic `Z2`

Further, we print:

`Lambda`	-	the optimal value of the smoothing hyperparameter
		divided by the sample size, i.e., \(\lambda/n\) in the notation
		of Komárek, Lesaffre and Hilton (2005)
`Log(Lambda)`	-	logarithm of the above
`df`	-	effective degrees of freedom of the model, see Section 2.2.3
		of Komárek, Lesaffre and Hilton (2005)
`AIC`	-	Akaike's information criterion of the model, see Section 2.2.3
		of Komárek, Lesaffre and Hilton (2005)

With argument spline set to TRUE, analogous table like that for the regression coefficients is printed also for the weights of the penalized Gaussian mixture (G-spline).

Usage

# S3 method for smoothSurvReg
print(x, spline, digits = min(options()$digits, 4), ...)
# S3 method for smoothSurvReg
summary(object, spline, digits = min(options()$digits, 4), ...)

Value

No return value, called to print the object.

Arguments

x: Object of class smoothSurvReg.
object: Object of class smoothSurvReg.
spline: TRUE/FALSE. If TRUE an information on fitted G-spline is printed.
digits: Controls the number of digits to print when printing numeric values. It is a suggestion only. Valid values are 1...22.
...: Further arguments passed to or from other methods.

Author

Arnošt Komárek arnost.komarek@mff.cuni.cz

References

Komárek, A., Lesaffre, E., and Hilton, J. F. (2005). Accelerated failure time model for arbitrarily censored data with smoothed error distribution. Journal of Computational and Graphical Statistics, 14, 726--745.