A nearestneighbors procedure is used in conjunction with the Epanechnikov kernel to define a kernel smooth of multinomial outcomes across the covariate space
1  smooth.patterns(dat, kfrac, bw)

dat 
The capturerecapture data in the form that is returned by

kfrac 
The approximate fraction of the data that is included in the support of the kernel for the local averages. 
bw 
A matrix a single column, with rownames that match the covariate
names in 
See Kurtz 2013, Chapter on multiple sclerosis
A list containing the original data (dat
), the smoothed data
(hpi
), and the effective sample sizes (ess
) for each local
average, or row, in the smoothed data
Zach Kurtz
Kurtz 2013
Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.
Please suggest features or report bugs with the GitHub issue tracker.
All documentation is copyright its authors; we didn't write any of that.