I?ve a data set and i want fit a kernel density estimate to the data. but using the k-nearest neighbour method. How i do this with R. thanks -- bertola at fastmail.fm --
Prof Brian Ripley
2003-Jun-10 17:57 UTC
[R] estimating a density by selecting the bandwidth
On Tue, 10 Jun 2003, Rafael Bertola wrote:> I?ve a data set and i want fit a kernel density estimate to the data. > but using the k-nearest neighbour method. > How i do this with R.Well, you define the exact algorithm you want to use, and then you program it, R being a fully-featured programming language. Your description is exceedingly vague: if you are hoping for an already implemented solution you might take a look at packge locfit (but that is only a rough fit to your `description'. -- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595