A Three-Step Method for Choosing the Number of Bootstrap Repetitions

By Donald W. K. Andrews, Moshe Buchinsky. 2000

319 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Finance, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/56c059cd4ee8c731fd3bc96f62eb178a503d4622

How researchers see this paper

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

In Economics and Econometrics

In Finance

In Statistics and Probability

Citations in the corpus

How Many Bootstrap Replicates Are Necessary? (2009, unknown): “Andrews and Buchinsky (1997, 2000, 2001) propose and evaluate a general three-step algorithm to specify B in the BS procedure.”
Aligning Learning Incentives of Students and Teachers: Results from a Social Experiment in Mexican High Schools (2015, unknown): “The critical assumptions are (1) the student can supply a minimal level of e¤ort at zero cost but must pay a xed (and variable) cost for supplying e¤ort above that minimal level; (2) the marginal product of teachers e¤ort is zero if the student chooses to supply only minimum e¤ort and (3) student ”
The Finite Sample Performance of Inference Methods for Propensity Score Matching and Weighting Estimators (2018, methodology): “In this context, we note that (Andrews and Buchinsky 2000) provide a method for choosing B to achieve a specific level of accuracy, measured as percentage deviation of a bootstrap quantity (e.g., a standard error or p-value) using a particular B from the ideal quantity under an infinite number of…”
Faster permutation inference in brain imaging (2016, background): “Various such criteria have been proposed (Andrews and Buchinsky, 2000; Davidson and MacKinnon, 2000; Fay and Follmann, 2002; Fay et al., 2007; Gandy, 2009; Kim, 2010; Sandve et al., 2011; Gandy and Rubin-Delanchy, 2013; Ruxton and Neuhäuser, 2013), and of particular interest is the interruption…”
Credit Constraints in the Market for Consumer Durables: Evidence from Micro Data on Car Loans (2008, unknown): “We did however experiment with the optimal number of bootstrap repetitions following Andrews and Buchinsky (2000).21 The results however did not change in any noteworthy degree in the cases we considered and we chose to report here only results based on 100 repetitions.”
The electricity consumption versus economic growth of the Polish economy (2012, methodology): “In our research we applied the recently developed procedure of establishing the number of bootstrap replications presented by Andrews and Buchinsky (2000).”
On measuring synchronization of bulls and bears: The case of East Asia (2008, methodology): “M is endogenously determined using the three step-method of Andrews and Buchinsky (2000) .1 A critical value is then obtained as the percent quantile, say W B crit , from the empirical distribution of the bootstrap test statistic.”
Sieve Wald and QLR Inferences on Semi/Nonparametric Conditional Moment Models (2014, unknown): “See, e.g., Andrews and Buchinsky (2000) for a thorough discussion about how to construct CS via bootstrap.”
Nonparametric estimation of distributional policy effects (2010, unknown): “Andrews and Buchinsky (2000) for a detailed discussion on how the choose the number of bootstrap replications B in applications.”
Testing for Co-integration in Vector Autoregressions with Non-Stationary Volatility (2010, unknown): “As is standard, the bootstrap samples are generated by imposing the null co-integration rank on the re-sampling scheme, thereby avoiding the di¢ culties with the use of unrestricted estimates of the impact matrix := 0; see Basawa et al. (1991) in the univariate case and Swensen (2006) in the multiva”
Matching and semi-parametric IV estimation, a distance-based measure of migration, and the wages of young men (2011, methodology): “Andrews and Buchinsky (2000, 2001) propose a three-step method for choosing the number of bootstrap repetitions.”
Generalized Method of Moments, Efficient Bootstrapping, and Improved Inference (2002, methodology): “For the purposes of this discussion, we ignore the error from simulation (see Andrews and Buchinsky 2000 for results on the size of that error).”
Cluster-Robust Bootstrap Inference in Quantile Regression Models (2014, methodology): “ssibly fractional) order moments of Z(˝) if the covariates have moments of slightly higher order: As long as p<q, the results in the appendix immediately give E j p n( P^ n(˝) ^ (˝))jp! EjZ(˝)jp. 3Andrews and Buchinsky (2000) andDavidson and MacKinnon ) provide methods for determining an appropri”
A Test for Superior Predictive Ability (2005, methodology): “This problem is closely related to the inconsistency of the bootstrap, when a parameter is on the boundary of the parameter space, as analyzed by Andrews (2000).”
Bootstrap Determination of the Co-Integration Rank in Heteroskedastic VAR Models (2014, methodology): “In practice, the cdf G ∗ r ,T (·) required in Step (iv) of Algorithm 1 will be unknown, but can be approximated in the usual way through numerical simulation; see, inter alia, Hansen (1996), Davidson and MacKinnon (2000) and Andrews and Buchinsky (2000). This is achieved by generating B (conditional”
How Large is the Bias is Self-Reported Disability? (2000, unknown): “To circumvent this problem we compute the small sample distribution of Tmax using a bootstrap procedure, using Andrews and Buchinsky’s (2000) recommendations for choosing the number of bootstrap repetitions.”