Segmentation of the mean of heteroscedastic data via cross-validation

نویسندگان

  • Sylvain Arlot
  • Alain Celisse
چکیده

This paper tackles the problem of detecting abrupt changes in the mean of a heteroscedastic signal by model selection, without knowledge on the variations of the noise. A new family of change-point detection procedures is proposed, showing that cross-validation methods can be successful in the heteroscedastic framework, whereas most existing procedures are not robust to heteroscedasticity. The robustness to heteroscedasticity of the proposed procedures is supported by an extensive simulation study, together with recent theoretical results. An application to Comparative Genomic Hybridization (CGH) data is provided, showing that robustness to heteroscedasticity can indeed be required for their analysis.

برای دانلود 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

Supplementary material for Segmentation of the mean of heteroscedastic data via cross-validation

In a previous version of this work [6, Chapter 7], Ĉ was defined as suggested in [7, 8], that is, Ĉ = 2K̂max.jump with the notation below. This yielded poor performances, which seemed related to the definition of Ĉ. Therefore, alternative definitions for Ĉ have been investigated, leading to the choice Ĉ = 2K̂thresh. throughout the paper, where K̂thresh. is defined by (2) below. The present appendi...

متن کامل

Heteroscedastic linear models for analysing process data

In this paper the guidelines for applying heteroscedastic linear models for analysing industrial process data is presented. Heteroscedastic linear models are considered as a good model family for the joint modelling of dispersion and mean. The model selection of heteroscedastic linear model is discussed considering the special features of industrial data. A procedure for dispersion model select...

متن کامل

Approximately unbiased estimation of conditional variance in heteroscedastic kernel ridge regression

In this paper we extend a form of kernel ridge regression for data characterised by a heteroscedastic noise process (introduced in Foxall et al. [1]) in order to provide approximately unbiased estimates of the conditional variance of the target distribution. This is achieved by the use of the leave-one-out cross-validation estimate of the conditional mean when fitting the model of the condition...

متن کامل

A REGION BASED IlVlAGE SEGlVlENTATION J\tlETHOD FOR J\tlULTI-CHANNEL DATA

become a tool in modern image analysis. Typical methods are highly non-linear, often ah-hoc and none so far have proved accessible to detailed statistical analysis even by asymptotic methods. Can such methods be any good? paper describes a segmentation procedure for multi-channel image data and attempts to develop an understanding of its statistical performance characteristics in terms of nonpa...

متن کامل

Choosing a penalty for model selection in heteroscedastic regression

Penalization is a classical approach to model selection. In short, penalization chooses the model minimizing the sum of the empirical risk (how well the model fits data) and of some measure of complexity of the model (called penalty); see FPE [1], AIC [2], Mallows’ Cp or CL [22]. A huge amount of literature exists about penalties proportional to the dimension of the model in regression, showing...

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

برای دانلود 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

عنوان ژورنال:
  • Statistics and Computing

دوره 21  شماره 

صفحات  -

تاریخ انتشار 2011