On the choice of parameters of change-point detection with application to stock exchange data  Cover Image

On the choice of parameters of change-point detection with application to stock exchange data
On the choice of parameters of change-point detection with application to stock exchange data

Author(s): Stanisław Jaworski, Konrad Furmańczyk
Subject(s): Economy
Published by: Szkoła Główna Gospodarstwa Wiejskiego w Warszawie
Keywords: V-Box Chart; mini-max rule; normal distribution

Summary/Abstract: Our paper is devoted to the study of V-Box Chart method in a parametric model. This algorithm is proposed to be used in the change-point detection in a sequence of observations. The choice of parameters in such an algorithm is heuristic. In our paper we use the mini-max rule for this choice and we control the probability that no signal is given, when the process is out of control as well as the probability of false alarm. We apply this algorithm to the detection of a change in stock exchange data.

  • Issue Year: XII/2011
  • Issue No: 1
  • Page Range: 87-96
  • Page Count: 10
  • Language: English