Abstract
For an estimator of locations to be useful for inferential purposes, a reliable method of estimating its variance is needed. In this paper, three methods of variance estimation are compared for ten location estimaters across a variety of parent distributions both symmetric and skew. Recommendations are made for different situations based on the bias of the variance estimater as well as the efficiency of the location estimater.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 221-248 |
| Number of pages | 28 |
| Journal | Communications in Statistics: Simulation and Computation |
| Volume | 10 |
| Issue number | 3 |
| DOIs | |
| State | Published - Jan 1 1981 |
Fingerprint
Keywords
- adaptive estimate
- and Phrases
- M estimate
- median
- Monte Carlo method
- trimmed mean
ASJC Scopus subject areas
- Modeling and Simulation
- Statistics and Probability
Cite this
Estimating the variances of robust estimators of location : Influence curve, jackknife and bootstrap. / Rocke, David M; Downs, George W.
In: Communications in Statistics: Simulation and Computation, Vol. 10, No. 3, 01.01.1981, p. 221-248.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Estimating the variances of robust estimators of location
T2 - Influence curve, jackknife and bootstrap
AU - Rocke, David M
AU - Downs, George W.
PY - 1981/1/1
Y1 - 1981/1/1
N2 - For an estimator of locations to be useful for inferential purposes, a reliable method of estimating its variance is needed. In this paper, three methods of variance estimation are compared for ten location estimaters across a variety of parent distributions both symmetric and skew. Recommendations are made for different situations based on the bias of the variance estimater as well as the efficiency of the location estimater.
AB - For an estimator of locations to be useful for inferential purposes, a reliable method of estimating its variance is needed. In this paper, three methods of variance estimation are compared for ten location estimaters across a variety of parent distributions both symmetric and skew. Recommendations are made for different situations based on the bias of the variance estimater as well as the efficiency of the location estimater.
KW - adaptive estimate
KW - and Phrases
KW - M estimate
KW - median
KW - Monte Carlo method
KW - trimmed mean
UR - http://www.scopus.com/inward/record.url?scp=0037638251&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0037638251&partnerID=8YFLogxK
U2 - 10.1080/03610918108812204
DO - 10.1080/03610918108812204
M3 - Article
AN - SCOPUS:0037638251
VL - 10
SP - 221
EP - 248
JO - Communications in Statistics Part B: Simulation and Computation
JF - Communications in Statistics Part B: Simulation and Computation
SN - 0361-0918
IS - 3
ER -