fits.compare: Numerical comparison of several fits

In RobustAFT: Truncated Maximum Likelihood Fit and Robust Accelerated Failure Time Regression for Gaussian and Log-Weibull Case

Numerical comparison of several fits

Description

Creates a class "fits.compare" object allowing the user to display model summary statistics in a form allowing easy comparison of models.

Usage

fits.compare(...)

Arguments

...	one or more class "lm", class "lm.robust" or class "TML" objects. Names given to objects in the list are used as labeling information in the printed output.

Details

The fits.compare function processes its arguments one at a time to create a named list of objects. The object returned is a member of class "fits.compare". Because of differences in the computed summary statistics, the list of input objects is currently limited to class "lm", class "lm.robust" and class "TML" objects. The print.fits.compare function displays a textual comparison of the input models, and the plot.fits.compare function provides comparative plots.

Value

An object of class "fits.compare" containing the list of input models to be compared.

See Also

TML.noncensored, plot.fits.compare

Examples

## Not run: 
     data(D243)
     Cost <- D243$Cost                             # Cost (Swiss francs)
     LOS  <- D243$LOS                              # Length of stay (days)
     Adm  <- D243$Typadm; Adm <- (Adm==" Urg")*1   # Type of admission 
                                                   # (0=on notification, 1=Emergency)

     lwrob <- TML.noncensored(log(Cost)~log(LOS)+Adm, errors="logWeibull")
     grob  <- TML.noncensored(log(Cost)~log(LOS)+Adm)
     reg   <- lm(log(Cost)~log(LOS)+Adm)

     fits.compare(least.squares=reg, TML.logWeibull=lwrob, TML.Gaussian=grob)

## End(Not run)

RobustAFT documentation built on Aug. 21, 2023, 5:13 p.m.