Detecting multiple change-points in the mean of Gaussian process by model selection

نویسنده

  • Emilie Lebarbier
چکیده

This paper deals with the problem of detecting the change-points in mean of a signal corrupted by an additive Gaussian noise. The number of changes and their positions are unknown. From a nonasymptotic point of view, we propose to estimate them with a method based on a penalized least-squares criterion. According to the results of Birgé and Massart, we choose the penalty function such that the resulting estimator minimizes the quadratic risk. This penalty depends on unknown constants and we propose a calibration leading to an automatic method. The performances of the method are assessed through simulation experiments. An application to real data is shown. Key-words: Detection of change-points; Penalized contrast; Model selection [email protected]. This research has been realized when E.Lebarbier was with Laboratoire de mathématiques Université Paris XI. Détection de ruptures dans la moyenne d'un processus gaussien par une méthode de sélection de modèle Résumé : Le papier traite du problème de détection de ruptures dans la moyenne d'un signal gaussien. Le nombre de ruptures et leurs localisations sont supposés inconnus. D'un point de vue non-asymptotique, nous proposons de les estimer à l'aide d'une méthode basée sur un critère des moindres carrés pénalisés. En appliquant les résultats de Birgé and Massart, nous choisissons une fonction de pénalité telle que l'estimateur pénalisé correspondant réalise le risque quadratic minimal. Cette pénalité dépend de constantes inconnues et nous proposons de les calibrer a n d'obtenir une automatique méthode. Une étude de simulations est menée pour évaluer la performance de la méthode, et une application sur des données réelles est réalisé. Mots-clés : Détection de ruptures ; Contraste pénalisé ; Sélection de modèles Change-Points in the Mean and Model Selection 3

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Bayesian Estimation of the Multiple Change Points in Gamma Process Using X-bar chart

The process personnel always seek the opportunity to improve the processes. One of the essential steps for process improvement is to quickly recognize the starting time or the change point of a process disturbance. Different from the traditional normally distributed assumption for a process, this study considers a process which follows a gamma process. In addition, we consider the possibility o...

متن کامل

Estimating Multiple Step Shifts in a Gaussian Process Mean with an Application to Phase I Control Chart Analysis

In preliminary analysis of control charts, one may encounter multiple shifts and/or outliers especially with a large number of observations. The following paper addresses this problem. A statistical model for detecting and estimating multiple change points in a finite batch of retrospective (phase I) data is proposed based on Likelihood Ratio Test. We consider a univariate normal distribution w...

متن کامل

S Estimating Multiple Step Shifts in a Gaussian Process Mean with an Application to Phase I Control Chart Analysis

In preliminary analysis of control charts, one may encounter multiple shifts and/or outliers especially with a large number of observations. The following paper addresses this problem. A statistical model for detecting and estimating multiple change points in a finite batch of retrospective (phase I) data is proposed based on Likelihood Ratio Test. We consider a univariate normal distribution w...

متن کامل

Bayesian Inference on Change Point Problems

Change point problems are referred to detect heterogeneity in temporal or spatial data. They have applications in many areas like DNA sequences, financial time series, signal processing, etc. A large number of techniques have been proposed to tackle the problems. One of the most difficult issues is estimating the number of the change points. As in other examples of model selection, the Bayesian...

متن کامل

Using CART to Detect Multiple Change Points in the Mean for large samples

A procedure is provided to detect multiple change-points in the mean of very large Gaussian signals. From an algorithmical point of view, visiting all possible con gurations of change-points cannot be performed on large samples. The proposed procedure runs CART rst in order to reduce the number of con gurations of change-points by keeping the relevant ones, and then runs an exhaustive search on...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • Signal Processing

دوره 85  شماره 

صفحات  -

تاریخ انتشار 2005