\name{drvkde}\alias{drvkde}\title{Kernel density derivative estimation}\description{  Compute kernel density derivative  estimates and standard errors for multivariate data. }\usage{drvkde(x, drv, bandwidth, gridsize, range.x, binned=FALSE, se=TRUE)}\arguments{  \item{x}{data matrix or matrix of binning counts}   \item{drv}{vector of derivative indices}  \item{bandwidth}{vector of bandwidths}  \item{gridsize}{vector of grid sizes}  \item{range.x}{list of vector of ranges for x}  \item{binned}{TRUE if x is binned counts or FALSE if x is    data matrix}   \item{se}{flag for computing the standard error of kernel    estimate} %%\item{estimate.positive}{flag for ensuring all estimate values are positive}}\value{  Returns a list with fields    \code{x.grid}  - grid points \cr  \code{est} - kernel estimate of partial derivative of density function  indicated by \code{drv}\cr  \code{se} -  estimate of standard error of \code{est} (if \code{se=TRUE}).}\details{The estimates and standard errors are computed over a grid of binned counts  \code{x.grid}. If the binned counts are not supplied then they are computed  inside this function.  If \code{gridsize} and \code{range.x} are not supplied, they are  computed inside this function.}\examples{## univariatex <- rnorm(100)fhat <- drvkde(x=x, drv=0, bandwidth=0.1)    ## KDE of ffhat1 <- drvkde(x=x, drv=1, bandwidth=0.1)   ## KDE of df/dx}\references{     Wand, M.P. and Jones, M.C. (1995) \emph{Kernel Smoothing}. Chapman	 \& Hall/CRC, London.}   \author{M.P. Wand}\keyword{smooth}