# Another look at bootstrapping the student t-statistic

• Published:

## Abstract

Let X, X 1, X 2, ... be a sequence of i.i.d. random variables with mean µ = EX. Let {v (n)1 , ..., v (n) n } n=1 be vectors of nonnegative random variables (weights), independent of the data sequence {X 1, ..., X n } n=1 , and put m n = Σ n i=1 v (n) i . Consider v (n)1 X 1, ..., v (n) n X n , a bootstrap sample, resulting from re-sampling or stochastically re-weighing the random sample X 1, ..., X n , n ≥ 1. Put $$\bar X_n = \sum\nolimits_{i = 1}^n {X_i } /n$$, the original sample mean, and define $$\bar X_{m_n }^* = \sum\nolimits_{i = 1}^n {v_i^{(n)} X_i /m_n }$$, where m n := Σ n i=1 v (n) i , the bootstrap sample mean. Thus, $$\bar X_{m_n }^* - \bar X_n = \sum\nolimits_{i = 1}^n {\left( {v_i^{(n)} /m_n - 1/n} \right)X_i }$$. Put V 2 n = Σ n i=1 (v (n) i /m n − 1/n)2 and let S 2 n , $$S_{m_n }^{*2}$$ respectively be the original sample variance and the bootstrap sample variance. The main aim of this exposition is to study the asymptotic behavior of the bootstrapped t-statistics $$T_{m_n }^* : = (\bar X_{m_n }^* - \bar X_n )/(S_n V_n )$$ and $$T_{m_n }^{**} : = \sqrt {m_n } (\bar X_{m_n }^* - \bar X_n )/S_{m_n }^*$$ in terms of conditioning on the weights via assuming that, as n → ∞, max1≤in (v (n) i /m n − 1/n)2/V 2 n = o(1) almost surely or in probability on the probability space of the weights. In consequence of these maximum negligibility conditions on the weights, a characterization of the validity of this approach to the bootstrap is obtained as a direct consequence of the Lindeberg-Feller central limit theorem (CLT). This view of justifying the validity of bootstrapping i.i.d. observables is believed to be new. The need for it arises naturally in practice when exploring the nature of information contained in a random sample via re-sampling, for example. Conditioning on the data is also revisited for Efron’s bootstrap weights under conditions on n, m n as n → ∞ that differ from requiring m n /n to be in the interval [λ 1, λ 2] with 0 < λ 1 < λ 2 < ∞ as in Mason and Shao (2001). The validity of the bootstrapped t-intervals is established for both approaches to conditioning.

This is a preview of subscription content, log in via an institution to check access.

## Subscribe and save

Springer+ Basic
\$34.99 /Month
• Get 10 units per month
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

## References

1. E. Arenal-Gutiérrez, C. Matrán, and J. A. Cuesta-Albertos, “On the Unconditional Strong Law of Large Numbers for the Bootstrap Mean”, Statist. Probab. Lett. 27, 49–60 (1996).

2. E. Arenal-Gutiérrez and C. Matrán, “A Zero-One Law Approach to the Central Limit Theorem for the Weighted Bootstrap Mean”, Ann. Probab. 24, 532–540 (1996).

3. K. B. Athreya, “Strong Law for the Bootstrap”, Statist. Probab. Lett. 1, 147–150 (1983).

4. M. Csörgő and M. M. Nasari, “Asymptotics of Randomly Weighted u- and v-Statistics: Application to Bootstrap”, J. Multivar. Anal. 121, 176–192 (2013).

5. S. Csörgő, “On the Law of Large Numbers for the Bootstrap Mean”, Statist. Probab. Lett. 14, 1–7 (1992).

6. S. Csörgő and A, Rosalsky, “A Survey of Limit Laws for Bootstrapped Sums”, International J. Math. and Math. Sci. 45, 2835–2861 (2003).

7. S. Csörgő and D. M. Mason, “Bootstrapping Empirical Functions”, Ann. Statist. 17, 1447–1471 (1989).

8. A. DasGupta, Asymptotic Theory of Statistics and Probability (Springer, New York, 2008).

9. B. Efron, “Bootstrap Methods: Another Look at the Jackknife”, Ann. Statist. 7, 1–26 (1979).

10. B. Efron and R. Tibshirani, An Introduction to the Bootstrap (Chapman & Hall, New York-London, 1993).

11. E. Giné, Lectures on Some Aspects of the Bootstrap. Ecole dÉté de Probabilités de Saint-Flour XXVI- 1996, Ed. by E. Giné, G. R. Grimmett, and L. Saloff-Coste in Lectures on Probability Theory and Statistics (1996).

12. E. Giné, F. Götze, and D. M. Mason, “When is the Student t-Statistic Asymptotically Normal?”, Ann. Probab. 25, 1514–1531 (1997).

13. E. Giné and J. Zinn, “Necessary Conditions for the Bootstrap of the Mean”, Ann. Statist. 17, 684–691 (1989).

14. J. Hájek, “Some Extensions of the Wald-Wolfowitz-Noether Theorem”, Ann. Math. Statist. 32, 506–523 (1961).

15. P. Hall, “On the Bootstrap and Confidence Intervals”, Ann. Statist. 14, 1431–1452 (1986).

16. D. M. Mason and M. A. Newton, “A Rank Statistics Approach to the Consistency of a General Bootstrap”, Ann. Statist. 20, 1611–1624 (1992).

17. D. M. Mason and Q. Shao, “Bootstrapping the Student t-Statistic”, Ann. Probab. 29, 1435–1450 (2001).

18. C. Morris, “Central Limit Theorems for Multinomial Sums”, Ann. Statist. 3, 165–188 (1975).

19. D. B. Rubin, “The Baysian Bootstrap”, Ann. Statist. 9, 130–134 (1981).

20. C. Weng, “On a Second-Order Asymptotic Property of the Bayesian Bootstrap Mean”, Ann. Statist. 17, 705–710 (1989).

## Author information

Authors

### Corresponding author

Correspondence to M. Csörgő.

Dedicated to the memory of Sándor Csörgő.

Csörgő, M., Martsynyuk, Y.V. & Nasari, M.M. Another look at bootstrapping the student t-statistic. Math. Meth. Stat. 23, 256–278 (2014). https://doi.org/10.3103/S1066530714040024

• Accepted:

• Published:

• Issue Date:

• DOI: https://doi.org/10.3103/S1066530714040024