Saudi Cultural Missions Theses & Dissertations
Permanent URI for this communityhttps://drepo.sdl.edu.sa/handle/20.500.14154/10
Browse
1 results
Search Results
Item Restricted Functional Linear Regression: Linear Hypothesis Testing With Functional Response(Saudi Digital, 2023-06-12) Alsaeed, Rasha; Yu, DengdengHypothesis testing is a crucial aspect of functional data analysis, allowing researchers to make inferential decisions based on samples of functional data. The inherent infinite dimensionality of functional data makes conventional hypothesis testing methods, such as Hotelling’s T2, difficult to apply due to the singularity of the sample covariance matrix. To address this issue, a common practice is to project functional data into a lower-dimensional space before conducting hypothesis tests. However, the choice of projection space can impact the validity and power of the tests, as the projected hypothesis problem may not be equivalent to the original problem. In this thesis, we propose a novel hypothesis testing procedure that establishes an optimal projection space where the original and projected hypothesis problems are equivalent and achieves the best power. The theoretical properties of the proposed test are systematically investigated, and a method for constructing the optimal projection space is provided. We demonstrate in this thesis that classical functional mean testing problems, including one-sample, two-sample, and multisample cases, and predictor significance testing can be reduced to special cases of the proposed method. To assess the performance of our proposed test, we conduct extensive simulations and analyze real data. Our results show the superiority of the proposed projection test for functional linear hypotheses in the function-on-scalar regression linear model. The findings of this thesis contribute to the advancement of hypothesis testing in functional data analysis by addressing the challenges posed by the infinite nature of functional data and providing a novel approach to establish an optimal projection space for improved hypothesis testing performance.23 0