Advisory Committee Chair
Ian W Knowles
Advisory Committee Members
Shangbing Ai
Marius N Nkashama
Roger B Sidje
Chengcui Zhang
Document Type
Dissertation
Date of Award
2023
Degree Name by School
Doctor of Philosophy (PhD) College of Arts and Sciences
Abstract
In this thesis, we forecast the trend and price of a selected stock using four new algorithms and a previous mathematical model built from economic ‘tendencies’. We use a new variational regularization method to determine the coefficients of the delay differential equation in the previous mathematical model. We also apply our new approach to obtaining initial points for the regularization method as well as a technique for mitigating the exponential growth model error throughout the minimization process of the regularization method. Lastly, we use the newly discovered coefficients together with a new iterative improvement procedure and a knowledge of the stock volatility to greatly improve the price prediction. We compare our model’s predictions to actual stock prices, forecasts from a previous model, and predictions from a deep learning model.
Recommended Citation
Cox, Kyoung Joo, "Functional Least Squares Minimization" (2023). All ETDs from UAB. 440.
https://digitalcommons.library.uab.edu/etd-collection/440