### Abstract

Tests on multivariate means that are hypothesized to be in a specified direction have received attention from both theoretical and applied points of view. One of the most common procedures used to test this cone alternative is the likelihood ratio test (LRT) assuming a multivariate normal model for the data. However, the resulting test for an ordered alternative is biased in that the only usable critical values are bounds on the null distribution. The present paper provides empirical evidence that bootstrapping the null distribution of the likelihood ratio statistic results in a bootstrap test (BT) with comparable power properties without the additional burden of assuming multivariate normality. Additionally, the tests based on the LRT statistic can reject the null hypothesis in favor of the alternative even though the true means are far from the alternative region. The BT also has similar properties for normal and nonnormal data. This anomalous behavior is due to the formulation of the null hypothesis and a possible remedy is to reformulate the null to be the complement of the alternative hypothesis. We discuss properties of a BT for the modified set of hypotheses (MBT) based on a simulation study. The resulting test is conservative in general and in some specific cases has power estimates comparable to those for existing methods. The BT has higher sensitivity but relatively lower specificity, whereas the MBT has higher specificity but relatively lower sensitivity.

Original language | English (US) |
---|---|

Pages (from-to) | 2302-2315 |

Number of pages | 14 |

Journal | Journal of Statistical Planning and Inference |

Volume | 137 |

Issue number | 7 |

DOIs | |

State | Published - Jul 1 2007 |

### Fingerprint

### Keywords

- Cone
- Correlated data
- Likelihood ratio
- Mean
- Nonparametric
- One-sided
- Ordered
- Positive orthant

### ASJC Scopus subject areas

- Statistics, Probability and Uncertainty
- Applied Mathematics
- Statistics and Probability

### Cite this

*Journal of Statistical Planning and Inference*,

*137*(7), 2302-2315. https://doi.org/10.1016/j.jspi.2006.07.011

**Bootstrap tests for multivariate directional alternatives.** / Minhajuddin, Abu T M; Frawley, William H.; Schucany, William R.; Woodward, Wayne A.

Research output: Contribution to journal › Article

*Journal of Statistical Planning and Inference*, vol. 137, no. 7, pp. 2302-2315. https://doi.org/10.1016/j.jspi.2006.07.011

}

TY - JOUR

T1 - Bootstrap tests for multivariate directional alternatives

AU - Minhajuddin, Abu T M

AU - Frawley, William H.

AU - Schucany, William R.

AU - Woodward, Wayne A.

PY - 2007/7/1

Y1 - 2007/7/1

N2 - Tests on multivariate means that are hypothesized to be in a specified direction have received attention from both theoretical and applied points of view. One of the most common procedures used to test this cone alternative is the likelihood ratio test (LRT) assuming a multivariate normal model for the data. However, the resulting test for an ordered alternative is biased in that the only usable critical values are bounds on the null distribution. The present paper provides empirical evidence that bootstrapping the null distribution of the likelihood ratio statistic results in a bootstrap test (BT) with comparable power properties without the additional burden of assuming multivariate normality. Additionally, the tests based on the LRT statistic can reject the null hypothesis in favor of the alternative even though the true means are far from the alternative region. The BT also has similar properties for normal and nonnormal data. This anomalous behavior is due to the formulation of the null hypothesis and a possible remedy is to reformulate the null to be the complement of the alternative hypothesis. We discuss properties of a BT for the modified set of hypotheses (MBT) based on a simulation study. The resulting test is conservative in general and in some specific cases has power estimates comparable to those for existing methods. The BT has higher sensitivity but relatively lower specificity, whereas the MBT has higher specificity but relatively lower sensitivity.

AB - Tests on multivariate means that are hypothesized to be in a specified direction have received attention from both theoretical and applied points of view. One of the most common procedures used to test this cone alternative is the likelihood ratio test (LRT) assuming a multivariate normal model for the data. However, the resulting test for an ordered alternative is biased in that the only usable critical values are bounds on the null distribution. The present paper provides empirical evidence that bootstrapping the null distribution of the likelihood ratio statistic results in a bootstrap test (BT) with comparable power properties without the additional burden of assuming multivariate normality. Additionally, the tests based on the LRT statistic can reject the null hypothesis in favor of the alternative even though the true means are far from the alternative region. The BT also has similar properties for normal and nonnormal data. This anomalous behavior is due to the formulation of the null hypothesis and a possible remedy is to reformulate the null to be the complement of the alternative hypothesis. We discuss properties of a BT for the modified set of hypotheses (MBT) based on a simulation study. The resulting test is conservative in general and in some specific cases has power estimates comparable to those for existing methods. The BT has higher sensitivity but relatively lower specificity, whereas the MBT has higher specificity but relatively lower sensitivity.

KW - Cone

KW - Correlated data

KW - Likelihood ratio

KW - Mean

KW - Nonparametric

KW - One-sided

KW - Ordered

KW - Positive orthant

UR - http://www.scopus.com/inward/record.url?scp=33947253383&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947253383&partnerID=8YFLogxK

U2 - 10.1016/j.jspi.2006.07.011

DO - 10.1016/j.jspi.2006.07.011

M3 - Article

AN - SCOPUS:33947253383

VL - 137

SP - 2302

EP - 2315

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

IS - 7

ER -