Affiliations: [a]
College of Agriculture, University of Sargodha, Sargodha, Pakistan
| [b]
Department of Mathematics and Statistics, University of Agriculture Faisalabad, Pakistan
| [c]
Department of Statistics, Faculty of Science, King Abdulaziz University, Saudi Arabia
| [d] Economics, FAST National University of Computer and Emerging Sciences, Chiniot Faisalabad Campus | [e]
Department of Statistics, University of Isfahan, Isfahan, Iran
| [f]
Department of Industrial Engineering, Faculty of Engineering-Rabigh, King Abdulaziz University, Jeddah, Saudi Arabia
Correspondence:
[*]
Corresponding author. Muhammad Aslam, Department of Statistics, Faculty of Science, King Abdulaziz University, Saudi Arabia. E-mail: [email protected].
Abstract: The application of classical statistical methods is not feasible given the presence of imprecise, fuzzy, uncertain, or undetermined observations in the underlying dataset. This is due to the existence of uncertainties pervading every aspect of real-life situations, which cannot always be accurately addressed by classical statistical approaches. In order to tackle this problem, a new methodology known as neutrosophic analysis of variance (NANOVA) has been developed as an extension of classical approaches to analyze datasets with uncertainty. The proposed approach can be applied regardless of the number of factors and replications. Moreover, NANOVA introduces a novel matrix-based approach to derive the F_N-test in an uncertain environment. To assess the effectiveness of NANOVA, various real datasets have been employed, and research findings on single- and two-factor NANOVAs with measures of indeterminacy have been presented. According to our comparisons, NANOVA provides a more informative, efficient, flexible, and reliable approach to deal with uncertainties than classical statistical methods. Therefore, there is a need to go beyond conventional statistical techniques and adopt advanced methodologies that can effectively handle uncertainties.
