Electronic Theses and Dissertations

Robert Vaughn

Date

2020

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Mathematical Sciences

Committee Chair

E. Olúṣẹ́gun George

Committee Member

E. Olúṣẹ́gun George

Committee Member

Su Chen

Committee Member

Dale Bowman

Abstract

Volatility of stocks is one of the most important factors in decision making by stock traders. It is usually manifested by large sudden fluctuations in stock prices and interspersed with short periods of price stability. It constitutes a critical factor in the decision to sell or buy, with the shrewd investor reaping great returns when the risk posed by volatility, high or low, is correctly harnessed in the decision making. As can be expected due to a common shock to the market such as COVID-19, volatilities of stocks within industry are statistically correlated. A shock in terms of price increase or decrease to one stock would usually have associated shock to other stocks in the same industry as traders move assets. A random effect facilitates joint modeling of stocks without concern for complicated covariance matrices that are associated with multivariate models. In this thesis we explore the use of random effects models to accommodate correlations of volatilities of stocks within industry. We introduce three different random effects models. Our analysis shows that our additive random effect model and multiplicative random effect model I works well to estimate the volatility and the correlation in the returns. We found a second multiplicative random effect model that is more restrictive to data that are correlated.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to ProQuest

Download

Share

COinS