On a wrapped kernel class of density estimators for circular data
The wrapped kernel density estimator studied in this work is a type of delta sequence density estimator whose delta function sequence is obtained by wrapping a scaled version of a probability density function defined on the real line. An asymptotic expansion for the mean integrated squared error (MISE) of this wrapped kernel density estimator is derived allowing for a direct comparison with other density estimators from the literature, such as the standard kernel density estimator and the Parzen-Rosenblatt type estimator. The wrapped Cauchy kernel density estimator will receive special attention.
Keywords: Circular data Density estimation Mean integrated squared error