A bootstrap test on the number of modes of a density
In this chapter we consider another application of bootstrap in nonparametric density estimation. We present results about the (random) number of modes of a kernel density estimate. These results are applied in the asymptotic treatment of bootstrap tests on the number of modes of a density. The material of this chapter is also contained in Mammen, Marron, and Fisher (1992).
Unable to display preview. Download preview PDF.