znuclear {bestglm} | R Documentation |
Data on 32 nuclear power plants. The response variable is cost and there are ten covariates.
data(znuclear)
A data frame with 32 observations on the following 12 variables. All quantitative variables, except date, have been logged and standardized to have mean 0 and variance 1.
date
T1
T2
capacity
PR
NE
CT
BW
N
PT
cost
Davison (2003) explores fitting models to this data using forward and backward stepwise regression. In this modelling logs of quantiative variables are used. We have also standardized this data to facilitate comparison with other techniques such as LARS and principal component regression.
Davison and Hinkley (1997, Example 6.8, 6.10, 6.12) use this data in a series of examples. Example 6.8: estimation of prediction error. Example 6.10: prediction error using cross-validation and bootstrapping. Example 6.12: subset model selection using cross-validation.
Obtained from the CRAN package boot.
Davison, A. C. (2003). Statistical Models. Cambridge: Cambridge University Press.
Davison, A.C. and Hinkley, D.V. (1997). Bootstrap Methods and their Application. Cambridge University Press.
data(znuclear) bestglm(znuclear, IC="BICq")