- 期刊
Testing Normality Based on Fuzzy Data
並列摘要
In this article, we first define a normal fuzzy random variable and then, using the α-pessimistic values of the imprecise observations, some methods for testing normality based on fuzzy observations are proposed. Using the popular test statistics, namely Kolmogorov-Smirnov, Cramer von Mises, Kuiper, Anderson-Darling, and Shapiro-Wilk, tests for normality of fuzzy data are constructed. By Monte Carlo simulations, the power values of the proposed tests are computed and then compared to each other under various alternatives. We then formulate some recommendations for the application of the studied tests in practice. Finally, we present an illustrative example which shows how the proposed procedures can be applied to test the goodness-of-fit of normality when a fuzzy sample is available.
延伸閱讀
- Cheng, Y. T. (2011). Fuzzy Canonical Discriminant Analysis. Journal of Data Analysis, 6(1), 115-133. https://doi.org/10.6338/JDA.201102_6(1).0008
- 郭信霖、李素惠（2013）。Study of Fuzzy Comprehensive Evaluation Based on Entropy。管理科學研究，()，43-48。https://doi.org/10.6873/MSR.201312/SP.0003
- Wu, P. L., Wu, M. C., & Chen, H. S. (2011). Applied Fuzzy Logic to Set Hypothesis of Sequential Test. Journal of Science and Innovation, 1(2), 29-38. https://doi.org/10.6654/JSI.201104_1(2).0004
- Martino, F. D., Perfilieva, I., & Sessa, S. (2015). Fuzzy Methods for Data Analysis. Advances in Fuzzy Systems, 2015(), 168-168. https://doi.org/10.1155/2015/957856
- Clara, N. (2007). Clustering Fuzzy Relational Data Base on Fuzzy Cardinality. International Journal of Fuzzy Systems, 9(4), 220-229. https://doi.org/10.30000/IJFS.200712.0005