### Abstract

In this paper, we study the robustness of Hotelling's T ^{2} when sampling from a mixture of two multivariate normal distributions. First we investigate the robustness of the T ^{2} statistic s test of the hypothesis H _{0}: μ = 0 when the underlying distribution is the mixture of two normal populations. The effect on the actual significance level when the T ^{2} test is used s a test of H _{0}: μ _{1} = 0 in the presence of exactly r outliers from N _{p}(μ _{2}, Σ) in a sample of size N is also considered.

Original language | English |
---|---|

Pages (from-to) | 89-103 |

Number of pages | 15 |

Journal | Random Operators and Stochastic Equations |

Volume | 10 |

Issue number | 1 |

Publication status | Published - Jan 1 2002 |

### Fingerprint

### All Science Journal Classification (ASJC) codes

- Analysis
- Statistics and Probability

### Cite this

^{2}when sampling from a contaminated normal rnodel: Robustness and outliers.

*Random Operators and Stochastic Equations*,

*10*(1), 89-103.

}

^{2}when sampling from a contaminated normal rnodel: Robustness and outliers',

*Random Operators and Stochastic Equations*, vol. 10, no. 1, pp. 89-103.

**Hotelling's T ^{2} when sampling from a contaminated normal rnodel : Robustness and outliers.** / Amey, A. K.A.; Gupta, A. K.

Research output: Contribution to journal › Article

TY - JOUR

T1 - Hotelling's T 2 when sampling from a contaminated normal rnodel

T2 - Robustness and outliers

AU - Amey, A. K.A.

AU - Gupta, A. K.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - In this paper, we study the robustness of Hotelling's T 2 when sampling from a mixture of two multivariate normal distributions. First we investigate the robustness of the T 2 statistic s test of the hypothesis H 0: μ = 0 when the underlying distribution is the mixture of two normal populations. The effect on the actual significance level when the T 2 test is used s a test of H 0: μ 1 = 0 in the presence of exactly r outliers from N p(μ 2, Σ) in a sample of size N is also considered.

AB - In this paper, we study the robustness of Hotelling's T 2 when sampling from a mixture of two multivariate normal distributions. First we investigate the robustness of the T 2 statistic s test of the hypothesis H 0: μ = 0 when the underlying distribution is the mixture of two normal populations. The effect on the actual significance level when the T 2 test is used s a test of H 0: μ 1 = 0 in the presence of exactly r outliers from N p(μ 2, Σ) in a sample of size N is also considered.

UR - http://www.scopus.com/inward/record.url?scp=29444439309&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=29444439309&partnerID=8YFLogxK

M3 - Article

VL - 10

SP - 89

EP - 103

JO - Random Operators and Stochastic Equations

JF - Random Operators and Stochastic Equations

SN - 0926-6364

IS - 1

ER -

^{2}when sampling from a contaminated normal rnodel: Robustness and outliers. Random Operators and Stochastic Equations. 2002 Jan 1;10(1):89-103.