Robust \(2^{k}\) factorial design with Weibull error distributions. (English) Zbl 1121.62300
Summary: It is well known that the least squares method is optimal only if the error distributions are normally distributed. However, in practice, non-normal distributions are more prevalent. If the error terms have a non-normal distribution, then the efficiency of least squares estimates and tests is very low. In this paper, we consider the \(2^{k}\) factorial design when the distribution of error terms are Weibull \(W(p,\sigma\)). From the methodology of modified likelihood, we develop robust and efficient estimators for the parameters in \(2^{k}\) factorial design. F statistics based on modified maximum likelihood estimators (MMLE) for testing the main effects and interaction are defined. They are shown to have high powers and better robustness properties as compared to the normal theory solutions. A real data set is analysed.
MSC:
| 62-XX | Statistics |
Keywords:
least squares; modified maximum likelihood; robustness; experimental design; Weibull distributionReferences:
| [1] | Bartlett M. S., Biometrika 40 pp 12– (1953) |
| [2] | DOI: 10.2307/2287904 · Zbl 0576.62042 · doi:10.2307/2287904 |
| [3] | Box G. E. P., Biometrika 40 pp 318– (1953) |
| [4] | Box G. E. P., Biometrika 51 pp 169– (1964) |
| [5] | Box G. E. P., Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building (1978) · Zbl 0394.62003 |
| [6] | Cohen A. C., Parameter Estimation in Reliability and Life Span Models (1988) · Zbl 0705.62093 |
| [7] | DOI: 10.1001/jama.211.1.69 · doi:10.1001/jama.211.1.69 |
| [8] | Fisher R. A., The Design of Experiments (1935) |
| [9] | Geary R. C., Biometrika 34 pp 209– (1947) |
| [10] | Hallinan A., Journal of Quality Technology 25 pp 85– (1993) |
| [11] | Hinkelmann K., Design and Analysis of Experiments, Volume 1 (1994) · Zbl 0805.62071 |
| [12] | DOI: 10.1002/0471725250 · Zbl 0536.62025 · doi:10.1002/0471725250 |
| [13] | DOI: 10.1081/STA-100104347 · Zbl 1008.62626 · doi:10.1081/STA-100104347 |
| [14] | Kendall M. G., The Advanced Theory of Statistics (1979) · Zbl 0416.62001 |
| [15] | Meeker W. Q., Statistical Methods for Reliability Data (1998) · Zbl 0949.62086 |
| [16] | Montgomery D. C., Design and Analysis of Experiments (1984) |
| [17] | Montgomery D. C., Introduction to Statistical Quality Control (1997) · Zbl 0997.62503 |
| [18] | DOI: 10.1002/9780470316795 · Zbl 0717.62089 · doi:10.1002/9780470316795 |
| [19] | Pearson E. S., Biometrika 23 pp 114– (1932) |
| [20] | DOI: 10.1016/0005-1098(86)90085-3 · Zbl 0589.93067 · doi:10.1016/0005-1098(86)90085-3 |
| [21] | DOI: 10.1093/biomet/72.1.67 · Zbl 0583.62026 · doi:10.1093/biomet/72.1.67 |
| [22] | Şenoğlu, B. 2000. ”Experimental design under nonnormality: skew distributions”. Ankara, Turkey: Department of Statistics, METU. PhD Dissertation, unpublished |
| [23] | DOI: 10.1081/STA-100104748 · Zbl 1008.62634 · doi:10.1081/STA-100104748 |
| [24] | DOI: 10.1002/1521-4036(200204)44:3<359::AID-BIMJ359>3.0.CO;2-K · doi:10.1002/1521-4036(200204)44:3<359::AID-BIMJ359>3.0.CO;2-K |
| [25] | Tan W. Y., Sampling Distributions In Terms of Laguerre Polynomials with Applications (1999) |
| [26] | Tiku M. L., Biometrika 54 pp 155– (1967) · doi:10.1093/biomet/54.1-2.155 |
| [27] | DOI: 10.1002/bimj.4710390802 · Zbl 0932.62079 · doi:10.1002/bimj.4710390802 |
| [28] | Tiku M. L., Robust Inference (1986) |
| [29] | DOI: 10.1080/03610929208830788 · Zbl 0800.62125 · doi:10.1080/03610929208830788 |
| [30] | DOI: 10.1016/S0895-7177(00)00119-9 · Zbl 0970.62036 · doi:10.1016/S0895-7177(00)00119-9 |
| [31] | DOI: 10.1081/STA-120002647 · Zbl 0991.62004 · doi:10.1081/STA-120002647 |
| [32] | Yates, F. 1937. ”Design and analysis of factorial experiments”. London: Imperial Bureau of Soil Sciences. Tech. Comm. No. 35 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
