Abstract
A sufficient statistic is a function of the observed data X containing all the information that X holds about the model. A complete sufficient statistic is one that reduces the data the most, without losing any information. More importantly, according to Rao–Blackwell-, Lehmann–Scheffé-theorems, statistical inference procedures must be based on such statistics for purposes of efficiency or optimality.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bahadur, R. R. (1954). Sufficiency and statistical decision functions. The Annals of Mathematical Statistics, 25, 423–462.
Barndorff-Nielsen, O. (1978). Information and exponential families in statistical theory. New York: Wiley.
Bickel, P. J., & Doksum, K. (2001). Mathematical statistics (2nd ed.). Englewood Cliffs, NJ: Prentice Hall.
Blackwell, D. (1947). Conditional expectation and unbiased sequential estimation. The Annals of Mathematical Statistics, 18(1), 105–110.
Brown, L. (1986). Fundamentals of statistical exponential families: With applications in statistical decision theory (Vol. 9). Hayward: Institute of Mathematical Statistics.
Diaconis, P. and Ylvisaker, D. (1979). Conjugate priors for exponential families. Ann. Statist. 7, 269–281.
Fisher, R. A. (1922). On the mathematical foundations of theoretical statistics. Philosophical Transactions of the Royal Society of London, 222, 310–366.
Fisher, R. A. (1934). Two new properties of mathematical likelihood Proceedings of the Royal Society A, 144, 285–307.
Fisher, R. A. (1958). Statistical methods for research workers (13th ed.). New York: Hafner
Halmos, P., & Savage, J. (1949). Application of the Radon-Nikodym theorem to the theory of sufficient statistics. The Annals of Mathematical Statistics, 20, 225–241.
Lehmann, E. (1959). Testing statistical hypothesis. New York: Wiley.
Lehmann, E. L., & Scheffé, H. (1947). On the problem of similar regions. Proceedings of the National Academy of Sciences of the United States of America, 33, 382–386.
Lehmann, E. L., & Scheffé, H. (1950). Completeness, similar regions, and unbiased estimation. I. Sankhya, 10, 305–340.
Lehmann, E. L., & Scheffé, H. (1955). Completeness, similar regions, and unbiased estimation. II. Sankhya, 15, 219–236.
Rao, C. R. (1945). Information and accuracy attainable in the estimation of statistical parameters. Bulletin of the Calcutta Mathematical Society, 37(3), 81–91.
Wasserman, L. (2003). All of statistics: A concise course in statistical inference. New York: Springer.
Author information
Authors and Affiliations
Appendix for Project: The Nonparametric Percentile Bootstrap of Efron
Appendix for Project: The Nonparametric Percentile Bootstrap of Efron
Let \(\hat{\sigma }_{n}\) be the standard error of \(\hat{\theta }_{n}\) (That is, \(\hat{\sigma }_{n}\) is an estimate of the standard deviation of \(\hat{\theta }_{n}\)). An asymptotic confidence interval of confidence level 1 −α for θ would follow from the relation \(P(z_{\alpha /2}\hat{\sigma }_{n} \leq \hat{\theta }_{n}-\theta \leq z_{1-\alpha /2}\hat{\sigma }_{n}) \approx 1-\alpha\), namely, it is the interval \([\hat{\theta }_{n} - z_{1-\alpha /2}\hat{\sigma }_{n},\hat{\theta }_{n} - z_{\alpha /2}\hat{\sigma }_{n}] = [\hat{\theta }_{n} - z_{\alpha /2}\hat{\sigma }_{n},\hat{\theta }_{n} + z_{1-\alpha /2}\hat{\sigma }_{n}] = [l,u]\), say. Now the bootstrap version \(\hat{\theta }_{n}^{{\ast}}\) of \(\hat{\theta }_{n}\) is, under the empirical \(\mathbf{P}^{{\ast}} =\widehat{ P}_{n}\), asymptotically Normal \(N(\hat{\theta }_{n},\hat{\sigma }_{n}^{2})\), so that the α∕2-th and (1 −α∕2)-th quantiles of \(\hat{\theta }_{n}^{{\ast}}\), q α∕2 ∗ and q 1−α∕2 ∗ say, are asymptotically equal to \(\hat{\theta }_{n} + z_{\alpha /2}\hat{\sigma }_{n} = l\) and \(\hat{\theta }_{n} + z_{1-\alpha /2}\hat{\sigma }_{n} = u\), respectively.
Hence the percentile bootstrap based confidence interval for θ is given by
Note that the construction of this interval only involves resampling from the data repeatedly to construct bootstrap versions \(\hat{\theta }_{n}^{{\ast}}\) of \(\hat{\theta }_{n}\); it does not involve the computation of the standard error \(\hat{\theta }_{n}\).
Although ( 4.62) does not involve computing the standard error \(\hat{\sigma }_{n}\), the latter is an important object in statistical analysis. It follows from the above that the variance \(\hat{\sigma }_{n}^{{\ast}2}\) of the \(\hat{\theta }_{n}^{{\ast}}\) values from the repeated resamplings provide an estimate of \(\hat{\sigma }_{n}^{2}\) [A rough estimate of \(\hat{\sigma }_{n}\) is also provided by \(([q_{1-\alpha /2}^{{\ast}}- q_{\alpha /2}^{{\ast}}]/2z_{1-\alpha /2})^{\frac{1} {2} }]\).
When the standard error \(\hat{\sigma }_{n}\) of \(\hat{\theta }_{n}\) is known in closed form, one may use the studentized or pivoted statistic \(T_{n} = (\hat{\theta }_{n}-\theta )/\hat{\sigma }_{n}\) which is asymptotically standard Normal N(0, 1). The usual CLT-based symmetric confidence interval for θ is given by
using P( | T n | ≤ z 1−α∕2) = 1 −α. The corresponding pivotal bootstrap confidence interval is based on the resampled values of \(T_{n}^{{\ast}} = (\hat{\theta }_{n}^{{\ast}}-\hat{\theta }_{n})/\hat{\sigma }_{n}^{{\ast}}\), where \(\hat{\sigma }_{n}^{{\ast}}\) is the bootstrap estimate of the standard error as described in the preceding paragraph. Let c α∕2 ∗ be such that P ∗( | T n ∗ | ≤ c α∕2 ∗) = 1 −α. The bootstrap pivotal confidence interval for θ is then
Suppose \(\hat{\theta }_{n}\) is based on i.i.d. observations \(X_{1},\ldots,X_{n}\), whose common distribution has a density (or a nonzero density component), and that it is a smooth function of sample means of a finite number of characteristics of X, or has a stochastic expansion (Taylor expansion) in terms of these sample means (such as the MLE in regular cases). It may then be shown that the coverage error of the CLT-based interval ( 4.63) is O(n −1), while that based on ( 4.64) is O(n −3∕2), a major advantage of the bootstrap procedure. The coverage error of the percentile interval ( 4.62) is O(n −1∕2), irrespective of whether the distribution of X is continuous or discrete.
Definition 4.6.
The coverage error of a confidence interval for a parameter θ is the (absolute) difference between the actual probability that the true parameter value belongs to the interval and the target level 1 −α.
Rights and permissions
Copyright information
© 2016 Springer-Verlag New York
About this chapter
Cite this chapter
Bhattacharya, R., Lin, L., Patrangenaru, V. (2016). Sufficient Statistics, Exponential Families, and Estimation. In: A Course in Mathematical Statistics and Large Sample Theory. Springer Texts in Statistics. Springer, New York, NY. https://doi.org/10.1007/978-1-4939-4032-5_4
Download citation
DOI: https://doi.org/10.1007/978-1-4939-4032-5_4
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4939-4030-1
Online ISBN: 978-1-4939-4032-5
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)