### Abstract

Fractional factorial design (FFD) has been widely used in engineering, agricultural, biology and social sciences. Although FFD is considered a conventional design, there still occurs a basic problem such as nonconstant variance may affect parameter estimation. Several parameter estimation methods that can be used are least squares, weighted least squares and more. The common method in estimation which is least squares method is applied initially to estimate the parameter. Then, the data was transformed to get the better result with generalized linear model. The estimator is finally calculated by applying robust method when it comes the problem of outlier to compare the performance of the model.

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

Title of host publication | 4th International Conference on Mathematical Sciences - Mathematical Sciences |

Subtitle of host publication | Championing the Way in a Problem Based and Data Driven Society, ICMS 2016 |

Publisher | American Institute of Physics Inc. |

Volume | 1830 |

ISBN (Electronic) | 9780735414983 |

DOIs | |

Publication status | Published - 27 Apr 2017 |

Event | 4th International Conference on Mathematical Sciences - Mathematical Sciences: Championing the Way in a Problem Based and Data Driven Society, ICMS 2016 - Putrajaya, Malaysia Duration: 15 Nov 2016 → 17 Nov 2016 |

### Other

Other | 4th International Conference on Mathematical Sciences - Mathematical Sciences: Championing the Way in a Problem Based and Data Driven Society, ICMS 2016 |
---|---|

Country | Malaysia |

City | Putrajaya |

Period | 15/11/16 → 17/11/16 |

### Fingerprint

### ASJC Scopus subject areas

- Physics and Astronomy(all)

### Cite this

*4th International Conference on Mathematical Sciences - Mathematical Sciences: Championing the Way in a Problem Based and Data Driven Society, ICMS 2016*(Vol. 1830). [080014] American Institute of Physics Inc.. https://doi.org/10.1063/1.4980998

**Statistical robust technique procedures for fractional factorial design with non constant variance.** / Yunus, Nadia Mohd; Muda, Nora.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*4th International Conference on Mathematical Sciences - Mathematical Sciences: Championing the Way in a Problem Based and Data Driven Society, ICMS 2016.*vol. 1830, 080014, American Institute of Physics Inc., 4th International Conference on Mathematical Sciences - Mathematical Sciences: Championing the Way in a Problem Based and Data Driven Society, ICMS 2016, Putrajaya, Malaysia, 15/11/16. https://doi.org/10.1063/1.4980998

}

TY - GEN

T1 - Statistical robust technique procedures for fractional factorial design with non constant variance

AU - Yunus, Nadia Mohd

AU - Muda, Nora

PY - 2017/4/27

Y1 - 2017/4/27

N2 - Fractional factorial design (FFD) has been widely used in engineering, agricultural, biology and social sciences. Although FFD is considered a conventional design, there still occurs a basic problem such as nonconstant variance may affect parameter estimation. Several parameter estimation methods that can be used are least squares, weighted least squares and more. The common method in estimation which is least squares method is applied initially to estimate the parameter. Then, the data was transformed to get the better result with generalized linear model. The estimator is finally calculated by applying robust method when it comes the problem of outlier to compare the performance of the model.

AB - Fractional factorial design (FFD) has been widely used in engineering, agricultural, biology and social sciences. Although FFD is considered a conventional design, there still occurs a basic problem such as nonconstant variance may affect parameter estimation. Several parameter estimation methods that can be used are least squares, weighted least squares and more. The common method in estimation which is least squares method is applied initially to estimate the parameter. Then, the data was transformed to get the better result with generalized linear model. The estimator is finally calculated by applying robust method when it comes the problem of outlier to compare the performance of the model.

UR - http://www.scopus.com/inward/record.url?scp=85019427222&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019427222&partnerID=8YFLogxK

U2 - 10.1063/1.4980998

DO - 10.1063/1.4980998

M3 - Conference contribution

AN - SCOPUS:85019427222

VL - 1830

BT - 4th International Conference on Mathematical Sciences - Mathematical Sciences

PB - American Institute of Physics Inc.

ER -