Presenter(s)

Rachel Sebastian

Comments

Presentation: 9:00-10:15 a.m., Kennedy Union Ballroom

Download

Download Project (309 KB)

Description

An important topic in finance is the question: What types of statistical models can characterize the distribution of stock returns? Stocks are partial shares of a company, and stock returns are a way to measure price changes and show the value of the company. While previous research considered the normal distribution, many articles have found this distribution does not accurately fit stock return data. In our study we investigate how well normal and alternative distributions fit stock return data from the S&P Index and Russell 2000. The alternative distributions explored were lognormal, Laplace, and Cauchy. Using quantile plots, alternative distributions, and measures of fit, we found that distributions other than the normal provide a better model for the indices tested. In addition, the best parameters for the alternative distributions vary depending on the measure of fit. Which distributions best characterize stock returns is an ongoing subject of study, but our findings suggest that non-normal distributions may provide a better model for the distribution of stock returns.

Publication Date

4-19-2023

Project Designation

Capstone Project

Primary Advisor

Matthew Wascher

Primary Advisor's Department

Mathematics

Keywords

Stander Symposium, College of Arts and Sciences

Institutional Learning Goals

Vocation

Modeling the Distribution of Stock Returns

Share

COinS