Abstract
The validity of a classifier depends on the precision of the error estimator used to estimate its true error. This paper considers the necessary sample size to achieve a given validity measure, namely RMS, for resubstitution and leave-one-out error estimators in the context of LDA. It provides bounds for the RMS between the true error and both the resubstitution and leave-one-out error estimators in terms of sample size and dimensionality. These bounds can be used to determine the minimum sample size in order to obtain a desired estimation accuracy, relative to RMS. To show how these results can be used in practice, a microar-ray classification problem is presented.
| Original language | English |
|---|---|
| Title of host publication | 2010 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2010 |
| DOIs | |
| Publication status | Published - 2010 |
| Externally published | Yes |
| Event | 2010 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2010 - Cold Spring Harbor, NY, United States Duration: Nov 10 2010 → Nov 12 2010 |
Other
| Other | 2010 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2010 |
|---|---|
| Country | United States |
| City | Cold Spring Harbor, NY |
| Period | 11/10/10 → 11/12/10 |
Fingerprint
ASJC Scopus subject areas
- Genetics
- Signal Processing
Cite this
RMS bounds and sample size considerations for error estimation in linear discriminant analysis. / Zollanvari, Amin; Braga-Neto, Ulisses M.; Dougherty, Edward R.
2010 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2010. 2010. 5719691.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - RMS bounds and sample size considerations for error estimation in linear discriminant analysis
AU - Zollanvari, Amin
AU - Braga-Neto, Ulisses M.
AU - Dougherty, Edward R.
PY - 2010
Y1 - 2010
N2 - The validity of a classifier depends on the precision of the error estimator used to estimate its true error. This paper considers the necessary sample size to achieve a given validity measure, namely RMS, for resubstitution and leave-one-out error estimators in the context of LDA. It provides bounds for the RMS between the true error and both the resubstitution and leave-one-out error estimators in terms of sample size and dimensionality. These bounds can be used to determine the minimum sample size in order to obtain a desired estimation accuracy, relative to RMS. To show how these results can be used in practice, a microar-ray classification problem is presented.
AB - The validity of a classifier depends on the precision of the error estimator used to estimate its true error. This paper considers the necessary sample size to achieve a given validity measure, namely RMS, for resubstitution and leave-one-out error estimators in the context of LDA. It provides bounds for the RMS between the true error and both the resubstitution and leave-one-out error estimators in terms of sample size and dimensionality. These bounds can be used to determine the minimum sample size in order to obtain a desired estimation accuracy, relative to RMS. To show how these results can be used in practice, a microar-ray classification problem is presented.
UR - http://www.scopus.com/inward/record.url?scp=79952791781&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79952791781&partnerID=8YFLogxK
U2 - 10.1109/GENSIPS.2010.5719691
DO - 10.1109/GENSIPS.2010.5719691
M3 - Conference contribution
SN - 9781612847924
BT - 2010 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS 2010
ER -