Sequential Cross-Validated Bandwidth Selection Under Dependence and Anscombe-Type Extensions to Random Time Horizons

  title={Sequential Cross-Validated Bandwidth Selection Under Dependence and Anscombe-Type Extensions to Random Time Horizons},
  author={Ansgar Steland},
  journal={Sequential Analysis},
  pages={326 - 350}
  • A. Steland
  • Published 30 May 2012
  • Mathematics
  • Sequential Analysis
Abstract To detect changes in the mean of a time series, one may use previsible detection procedures based on nonparametric kernel prediction smoothers that cover various classic detection statistics as special cases. Bandwidth selection, particularly in a data-adaptive way, is a serious issue and not well studied for detection problems. To ensure data adaptation, we select the bandwidth by cross-validation but in a sequential way leading to a functional estimation approach. This article… 
Detection of changes in dependent processes : Learning from algorithms , simulations and stochastic inference
Combining stochastic limit theory, simulation and applications provide a powerful approach to learn from data and obtain asymptotic results that hold true for rich nonparametric classes of time series.
Vertically Weighted Averages in Hilbert Spaces and Applications to Imaging: Fixed-Sample Asymptotics and Efficient Sequential Two-Stage Estimation
Abstract We discuss fixed-sample asymptotics and jackknife variance estimation for vertically weighted averages and the construction of related sequential two-stage confidence intervals. Those


Sequential Data-Adaptive Bandwidth Selection by Cross-Validation for Nonparametric Prediction
  • A. Steland
  • Mathematics, Computer Science
    Commun. Stat. Simul. Comput.
  • 2012
This work investigates the problem of bandwidth selection by cross-validation from a sequential point of view in a nonparametric regression model for sequential kernel smoothers in order to base these tasks on a single statistic.
Optimal sequential kernel detection for dependent processes
A new theorem on the convergence of the normed delay of the associated sequential detection procedure which holds for dependent time series under a weak mixing condition is proved and two new theorems about the existence and an explicit representation of optimal kernels minimizing the asymptotic normeddelay are provided.
Bandwidth Selection for Changepoint Estimation in Nonparametric Regression
A bootstrap procedure for selecting the bandwidth parameters in a nonparametric two-step estimation method results in a fully data-driven procedure for estimating a finite (but possibly unknown) number of changepoints in a regression function.
Asymptotic results for stopping times based on u-statistics
This paper considers a general stopping time based on U-statistics. The purpose is todevelop a general setting that will apply to many situations where the family ofdistributions from which we
A surveillance procedure for random walks based on local linear estimation
We study the problem of detecting a change in the trend of time series whose stochastic part behaves as a random walk. Interesting applications in finance and engineering come to mind. Local linear
Monitoring shifts in mean: Asymptotic normality of stopping times
Abstract We consider a sequential procedure designed to detect a possible change in the mean of a random sequence. The procedure is motivated by the problem of detecting an early change in the mean
Functional Limit Theorem for the Empirical Process of a Class of Bernoulli Shifts with Long Memory
We prove a functional central limit theorem for the empirical process of a stationary process Xt=Yt+Vt, where Yt is a long memory moving average in i.i.d. r.v.’s ζs, s ≤ t, and Vt=V (ζt, ζt-1,...) is
Estimation of a function with discontinuities via local polynomial fit with an adaptive window choice
We propose a method of adaptive estimation of a regression function and which is near optimal in the classical sense of the mean integrated error At the same time the estimator is shown to be very
Discontinuous versus smooth regression
Given measurements (x i , y i ), i = 1 n, we discuss methods to assess whether an underlying regression function is smooth (continuous or differentiable) or whether it has discontinuities. The