Bias reduction by taylor series<sup>∗</sup> (Q3765026)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4022362
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Bias reduction by taylor series<sup>∗</sup> |
scientific article; zbMATH DE number 4022362 |
Statements
Bias reduction by taylor series<sup>∗</sup> (English)
0 references
1987
0 references
bias reduction
0 references
Taylor series
0 references
delta method
0 references
infinitesimal Jackknife
0 references
unbiased estimates
0 references
0.805599570274353
0 references
0.8017321825027466
0 references
0.775879979133606
0 references
0.775416910648346
0 references
We consider the problem of estimating a function say t(\(\theta)\), given an estimate \({\hat \theta}\) with distribution determined by the unknown vector \(\theta\). Typically t(\({\hat \theta}\)) has bias \(0(n^{-1})\), written \(\sim n^{-1}\), and requires \(\sim n\) calculations, where n is the sample size (or minimum sample size for more than one sample). For a wide class of estimates \({\hat \theta}\) and any given k, we show how to construct an estimate of t(\(\theta)\) with bias \(\sim n^{-k}\) which still requires only \(\sim n\) calculations. For \(k\leq 4\) an explicit formula is given. The method can be extended to give unbiased estimates (UEs) when their form as a function of n is known.
0 references