Master’s Thesis at the University of Basrah Examines the Hybridization of Kriging with Adaptive LASSO
A master’s thesis at the College of Administration and Economics, University of Basrah, examined the hybridization of the multi-task Kriging model with adaptive LASSO regression to improve the accuracy of spatial estimation for high-dimensional data.
The thesis aimed to develop a hybrid statistical model that enhances spatial prediction efficiency and stability compared with conventional models.
Submitted by student Abdullah Salem Kaseb, the thesis examined Kriging models and variable selection techniques, combining them using the Least Angle Regression (LARS) algorithm.
The results demonstrated the hybrid model’s superior performance and effectiveness in both simulation and practical applications using salinity data from the Shatt Al-Arab River.
The thesis recommended adopting the proposed model and expanding the range of variables used in future studies.
Department of Media and Government Communication
