Downloads & Free Reading Options - Results

In this paper, the detection and estimation of change points of local parameters are studied by means of localization procedures and rank statistics. These techniques are also applied to detection and estimation of the change points of scale parameters and that of location parameters of directional data. Change point problem arises in many fields and attracts the attention of many authors. The techniques employed to detect and estimate the change point can generally be classified into two categories: parametric, and nonparametric. Bayesian methods also plays a major role. In this paper, the authors concentrate their attention on the problem of detection of change points of location parameter by localization and rank statistics when data are large. Their method is different from, and has some advantages over, the existing methods, such as CUSUM (cumulative sum) and Csorgo and Horvath's non-sequential nonparametric AMOC (at most one change) procedures. First, localized procedures reduce computation. Second, these detecting and estimating procedures require no moment condition, instead it is only assumed that observed data come from a continuous distribution with a unique median.