The Jackknife and the Bootstrap for General Stationary Observations

By Hans R. Künsch. 1989

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Invariant Causal Prediction for Nonlinear Models (2017, methodology): “The conﬁdence bands ˆ F , required for the computation of (cid:91) ACE (˜ x, x ), are obtained by a time series bootstrap (K¨unsch, 1989) as the fertility data contain temporal dependencies.”
Trend analysis of climate time series: A review of methods (2019, methodology): “Künsch (1989) presented the MBB (Fig.”
Higher‐Order Improvements of a Computationally Attractive k‐Step Bootstrap for Extremum Estimators (2002, background): “The former is often called the Kunsch (1989) blocking scheme, and the latter, the Carlstein (1986) scheme, although Hall (1985) considers both of these schemes in a related context.”
Neural network ensemble operators for time series forecasting (2014, methodology): “In each pair of ensembles, the ﬁrst is a training ensemble, combining multiple training initialisations and the second is based on bagging, bootstrapped as described by Kunsch (1989).”
Inference Functions and Quadratic Score Tests (2003, methodology): “We applied the block bootstrap strategy here (Künsch, 1989; Lahiri, 1996).”
Jackknife estimation of stationary autoregressive models (2013, unknown): “In the context of stationary time series Carlstein (1986) proposed an estimator of variance based on non-overlapping blocks while Künsch (1989) considered both jackknife and bootstrap methods of estimating standard errors by deleting whole blocks of observations.”
A nonparametric plug-in rule for selecting optimal block lengths for block bootstrap methods ☆ (2007, methodology): “Di erent versions of the block bootstrap method for dependent data have been put forward by Hall (1985), Carlstein (1986), K unsch (1989), Liu and Singh (1992), Politis and Romano (1992), Carlstein et al. (1998), and Paparoditis and Politis (2001), among others.”
Covariance matrix estimation and linear process bootstrap for multivariate time series of possibly increasing dimension (2015, methodology): “…bootstrap procedures in time series analysis, we mention the autoregressive (AR) sieve bootstrap [cf. Kreiss (1992, 1999), Bühlmann (1997), Kreiss, Paparoditis and Politis (2011)] and block bootstrap and its variations; cf. Künsch (1989), Liu and Singh (1992), Politis and Romano (1992, 1994), etc.”
A more general central limit theorem for m-dependent random variables with unbounded m (2000, methodology): “The moving blocks bootstrap was introduced by K unsch (1989) and Liu and Singh (1992) as an extension of the classical bootstrap by Efron (1979).”
A comparison of AdaBoost algorithms for time series forecast combination (2016, methodology): “In this study we apply Bagging to base learners of MLP and CART, using the moving block bootstrap (Kunsch 1989) which samples the original time series while preserving the temporal covariance structure”
More accurate, calibrated bootstrap confidence intervals for estimating the correlation between two time series (2014, methodology): “2.1 Pairwise Moving Block Bootstrap Resampling The pairwise moving block bootstrap (MBB) resampling technique, advocated by Mudelsee (2010), is used to resample blocks of data from the original time series {x(i), y(i)}ni=1 to form resamples that preserve the properties of the data-generating process”