Home > Research > Publications & Outputs > On the use of cross-validation to assess perfor...

Research

Links

Text available via DOI:

Keywords

View graph of relations

On the use of cross-validation to assess performance in multivariate prediction

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

On the use of cross-validation to assess performance in multivariate prediction. / Jonathan, P.; Krzanowski, W.J.; McCarthy, W.V.
In: Statistics and Computing, Vol. 10, No. 3, 2000, p. 209-229.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Jonathan, P, Krzanowski, WJ & McCarthy, WV 2000, 'On the use of cross-validation to assess performance in multivariate prediction', Statistics and Computing, vol. 10, no. 3, pp. 209-229. https://doi.org/10.1023/A:1008987426876

APA

Jonathan, P., Krzanowski, W. J., & McCarthy, W. V. (2000). On the use of cross-validation to assess performance in multivariate prediction. Statistics and Computing, 10(3), 209-229. https://doi.org/10.1023/A:1008987426876

Vancouver

Jonathan P, Krzanowski WJ, McCarthy WV. On the use of cross-validation to assess performance in multivariate prediction. Statistics and Computing. 2000;10(3):209-229. doi: 10.1023/A:1008987426876

Author

Jonathan, P. ; Krzanowski, W.J. ; McCarthy, W.V. / On the use of cross-validation to assess performance in multivariate prediction. In: Statistics and Computing. 2000 ; Vol. 10, No. 3. pp. 209-229.

Bibtex

@article{f189bcba26494c57b0048ab1d754ec29,
title = "On the use of cross-validation to assess performance in multivariate prediction",
abstract = "We describe a Monte Carlo investigation of a number of variants of cross-validation for the assessment of performance of predictive models, including different values of k in leave-k-out cross-validation, and implementation either in a one-deep or a two-deep fashion. We assume an underlying linear model that is being fitted using either ridge regression or partial least squares, and vary a number of design factors such as sample size n relative to number of variables p, and error variance. The investigation encompasses both the non-singular (i.e. n > p) and the singular (i.e. n ≤ p) cases. The latter is now common in areas such as chemometrics but has as yet received little rigorous investigation. Results of the experiments enable us to reach some definite conclusions and to make some practical recommendations.",
keywords = "Assessment of predictive models, Cross-validation, Partial least squares, Prediction, Ridge regression",
author = "P. Jonathan and W.J. Krzanowski and W.V. McCarthy",
year = "2000",
doi = "10.1023/A:1008987426876",
language = "English",
volume = "10",
pages = "209--229",
journal = "Statistics and Computing",
issn = "0960-3174",
publisher = "Springer Netherlands",
number = "3",

}

RIS

TY - JOUR

T1 - On the use of cross-validation to assess performance in multivariate prediction

AU - Jonathan, P.

AU - Krzanowski, W.J.

AU - McCarthy, W.V.

PY - 2000

Y1 - 2000

N2 - We describe a Monte Carlo investigation of a number of variants of cross-validation for the assessment of performance of predictive models, including different values of k in leave-k-out cross-validation, and implementation either in a one-deep or a two-deep fashion. We assume an underlying linear model that is being fitted using either ridge regression or partial least squares, and vary a number of design factors such as sample size n relative to number of variables p, and error variance. The investigation encompasses both the non-singular (i.e. n > p) and the singular (i.e. n ≤ p) cases. The latter is now common in areas such as chemometrics but has as yet received little rigorous investigation. Results of the experiments enable us to reach some definite conclusions and to make some practical recommendations.

AB - We describe a Monte Carlo investigation of a number of variants of cross-validation for the assessment of performance of predictive models, including different values of k in leave-k-out cross-validation, and implementation either in a one-deep or a two-deep fashion. We assume an underlying linear model that is being fitted using either ridge regression or partial least squares, and vary a number of design factors such as sample size n relative to number of variables p, and error variance. The investigation encompasses both the non-singular (i.e. n > p) and the singular (i.e. n ≤ p) cases. The latter is now common in areas such as chemometrics but has as yet received little rigorous investigation. Results of the experiments enable us to reach some definite conclusions and to make some practical recommendations.

KW - Assessment of predictive models

KW - Cross-validation

KW - Partial least squares

KW - Prediction

KW - Ridge regression

U2 - 10.1023/A:1008987426876

DO - 10.1023/A:1008987426876

M3 - Journal article

VL - 10

SP - 209

EP - 229

JO - Statistics and Computing

JF - Statistics and Computing

SN - 0960-3174

IS - 3

ER -