Bandwidth selection for kernel conditional density estimation. (Q5941105): Difference between revisions

We consider bandwidth selection for kernel estimators of conditional densities with one explanatory variable. Several bandwidth selection methods are derived, ranging from fast rules-of-thumb which assume the underlying densities are known to relatively slow procedures which use the bootstrap. The methods are compared and a practical bandwidth selection strategy which combines the methods is proposed. The methods are compared using two simulation studies and a real data set.