Publication details
Bandwidth matrix selectors for multivariate kernel density estimation
| Authors | |
|---|---|
| Year of publication | 2013 |
| Type | Conference abstract |
| MU Faculty or unit | |
| Citation | |
| Description | Multivariate kernel density estimators have become a common tool for empirical studies. These estimators depend on the choice of a crucial parameter -- the bandwidth matrix. In this paper, we propose an extension of the iterative method for finding the bandwidth matrix not only for density estimation but also for estimating its derivatives. This procedure is illustrated by a simulation study and two real data sets. |