expectreg (version 0.53)

methods: Methods for expectile regression objects

Description

Methods for objects returned by expectile regression functions.

Usage

# S3 method for expectreg
print(x, ...)
# S3 method for expectreg
summary(object,...)
# S3 method for expectreg
predict(object, newdata = NULL, with_intercept = T, ...)
# S3 method for expectreg
[(x, i)
# S3 method for expectreg
residuals(object, ...)
# S3 method for expectreg
resid(object, ...)
# S3 method for expectreg
fitted(object, ...)
# S3 method for expectreg
fitted.values(object, ...)
# S3 method for expectreg
effects(object, ...)
# S3 method for expectreg
coef(object, ...)
# S3 method for expectreg
coefficients(object, ...)
# S3 method for expectreg
confint(object, parm = NULL, level = 0.95, ...)

Value

[ returns a new object of class expectreg with a subset of covariates from the original fit.

resid returns the residuals in order of the response.

fitted returns the overall fitted values \(\hat{y}\) while effects returns the values for each covariate in a list.

coef returns a list of all regression coefficients separately for each covariate.

Arguments

x,object

An object of class expectreg as returned e.g. by the function expectreg.ls.

newdata

Optionally, a data frame in which to look for variables with which to predict.

with_intercept

Should the intercept be added to the prediction of splines?

i

Covariate numbers to be kept in subset.

level

Coverage probability of the generated confidence intervals.

parm

Optionally the confidence intervals may be restricted to certain covariates, to be named in a vector. Otherwise the confidence intervals for the fit are returned.

...

additional arguments passed over.

Author

Fabian Otto- Sobotka
Carl von Ossietzky University Oldenburg
https://uol.de

Elmar Spiegel
Georg August University Goettingen https://www.uni-goettingen.de

Details

These functions can be used to extract details from fitted models. print shows a dense representation of the model fit.

[ can be used to define a new object with a subset of covariates from the original fit.

The function coef extracts the regression coefficients for each covariate listed separately. For the function expectreg.boost this is not possible.

References

Schnabel S and Eilers P (2009) Optimal expectile smoothing Computational Statistics and Data Analysis, 53:4168-4177

Sobotka F and Kneib T (2010) Geoadditive Expectile Regression Computational Statistics and Data Analysis, doi: 10.1016/j.csda.2010.11.015.

See Also

expectreg.ls, expectreg.boost, expectreg.qp

Examples

data(dutchboys)

expreg <- expectreg.ls(hgt ~ rb(age,"pspline"),data=dutchboys,smooth="f",
                       expectiles=c(0.05,0.2,0.8,0.95))

print(expreg)

coef(expreg)

new.d = dutchboys[1:10,]
new.d[,2] = 1:10

predict(expreg,newdata=new.d)