White, Luther W2019-05-012019-05-012010https://hdl.handle.net/11244/319435In this thesis we use the information theoretic approach in selecting the bestmodel among many candidate models. It is shown that the information theoreticapproach is better than the standard R2 approach in selecting models. We useAkaike Information Criteria (AIC) to select the best model for resilient modulusof a soil and for a girder. This approach is applied to statistical models, neuralnetwork models and physics based models. The information theory approachis compared with the R2 approach and it is found that the information theo-retic approach is more stable and gives better results. The notion of rankingstability is introduced and is used as one of the reasons that makes informationtheory approach better than the R2 approach. Important results are capturedand compared to the results of the R2 method in two dierent data sets.x110 pagesapplication.pdfInformation theoryMathematical modelsMathematical statisticstext