Skip to contents

se is a generic function which extracts standard errors from objects.

Usage

se(x)

Arguments

Object Input

Object to interrogate for standard errors.

x

an object for which the extraction of standard errors is meaningful.

Value

Standard errors extracted from the model object x.

Details

This function extracts stored standard errors; it does not select bandwidths, refit a model, or run a bootstrap. For estimators with a se fitting control, request se = TRUE when those errors will be needed. With se = FALSE, errors are deliberately absent and this accessor explains how to refit with se = TRUE, reusing the bandwidth object without repeating bandwidth selection. Requested but unavailable standard errors may remain NA; extraction does not replace or recompute them.

See also

fitted, residuals, coef, and gradients, for related methods; np for supported objects.

Author

Tristen Hayfield tristen.hayfield@gmail.com, Jeffrey S. Racine racinej@mcmaster.ca

Note

This method currently only supports objects from the np library.

Examples

x <- rnorm(10)
se(npudens(npudensbw(~x), se = TRUE))
#>  [1] 0.16043480 0.07763219 0.07913694 0.15456448 0.13887398 0.15769815
#>  [7] 0.11835344 0.15505797 0.15871814 0.14466607