# Estimation of the location of the cusp of a continuous density

Prakasa Rao, B. L. S. (1968) Estimation of the location of the cusp of a continuous density Annals of Mathematical Statistics, 39 (1). pp. 76-87. ISSN 0003-4851

Full text not available from this repository.

Related URL: http://dx.doi.org/10.1214/aoms/1177698506

## Abstract

Chernoff and Rubin  and Rubin  investigated the problem of estimation of the location of a discontinuity in density. They have shown that the maximum likelihood estimator (MLE) is hyper-efficient under some regularity conditions on the density and that asymptotically the estimation problem is equivalent to that of a non-stationary process with unknown center of non-stationarity. We have obtained here similar results for a family of densities f(X, ϑ) which are continuous with a cusp at the point ϑ. In this connection, it is worth noting that Daniels  has obtained a modified MLE for the family of densities f(X, ϑ) = C(λ) exp {−|X − ϑ|λ}, for λ such that ½ < λ ½ 1, where C(λ) is a constant depending on λ and he has shown that this estimator is asymptotically efficient. In this paper, we shall show that the MLE of ϑ is hyper-efficient for the family of densities f(X, ϑ) given by (1.1) log f(X,ϑ) = ε(X,ϑ)|X − ϑ)|X−ϑ| λ + g(X,ϑ) for |X| ≤ A = g(X,ϑ) for |X| > A where A is a constant greater than zero, (1.2) ε(X,ϑ) = β(ϑ) if X < ϑ = γ(ϑ) if X > ϑ, (1.3) 0 < λ < ½ and ϑ ε(a, b) where -A < a < b < A, under some regularity conditions on f(X, ϑ) and we shall derive the asymptotic distribution of the MLE implicitly. Section 2 contains the regularity conditions imposed on the family of densities f(X, ϑ). Section 3 contains some results related to the asymptotic properties of the MLE. The estimation problem is reduced to that of a stochastic process in Section 4. The asymptotic distribution of MLE is obtained in Section 5.

Item Type: Article Copyright of this article belongs to Institute of Mathematical Statistics. 73739 07 Dec 2011 05:34 07 Dec 2011 05:34

Repository Staff Only: item control page