Adobe PDF (354.75 kB)
Title Details:
Cramér-Rao Inequality, Fisher Information, and Efficient Estimators
Authors: Kourouklis, Stavros
Petropoulos, Konstantinos
Piperigkou, Violetta
Reviewer: Batsidis, Apostolos
Description:
Abstract:
One way to circumvent the problem of the non-existence of an optimal estimator is to restrict our search to a UMVUE (Uniformly Minimum Variance Unbiased Estimator). Recall that a UMVUE has the minimum variance among all unbiased estimators of the quantity g(θ). In this chapter, we deal with the existence of a lower bound for the variance of a statistic. Then, we see that in certain cases, this lower bound can be used to show that an estimator is a UMVUE, with variance equal to the lower bound. Such an estimator is referred to as an efficient estimator. However, it should be noted that the main application and use of the lower bound lies in the asymptotic properties of estimators.
Linguistic Editors: Gyftopoulou, Ourania
Type: Chapter
Creation Date: 2015
Item Details:
License: http://creativecommons.org/licenses/by-nc-nd/3.0/gr
Handle http://hdl.handle.net/11419/5689
Bibliographic Reference: Kourouklis, S., Petropoulos, K., & Piperigkou, V. (2015). Cramér-Rao Inequality, Fisher Information, and Efficient Estimators [Chapter]. In Kourouklis, S., Petropoulos, K., & Piperigkou, V. 2015. Topics in Parametric Statistical Inference: estimation and confidence intervals [Undergraduate textbook]. Kallipos, Open Academic Editions. https://hdl.handle.net/11419/5689
Language: Greek
Is Part of: Topics in Parametric Statistical Inference: estimation and confidence intervals
Publication Origin: Kallipos, Open Academic Editions