Dissertation/Thesis Abstract

The Logical Approach to Equity Returns
by Liu, Tan, M.S., California State University, Long Beach, 2019, 51; 22587243
Abstract (Summary)

The returns to equities must be narrowly defined as real money that has been returned to investors. It cannot be an idea of money that can only be realized under certain conditions.

Capital gains cannot be considered returns to investors because they are zero-sum by design and negative-sum in practice with fees. Unrealized capital gain is not real money. It is an abstract idea that cannot be realized by all investors in practice because of liquidity constraints. Realized capital gains are not returns to investors because the realized profits for earlier investors typically come from the inflow of money from new investors. When one investor experiences a profit, another is also experiencing a loss.

Stock buybacks cannot be considered returns because of dilution. Many companies that buy back stocks also issue additional shares before or after the buyback. Those that have not engaged in false buybacks still have the option to dilute in the future and nullify what they returned to investors. Only dividends can be considered real returns to investors because it is money paid to investors by the underlying businesses, and it cannot be rescinded after it is paid.

The analysis focused on dividend companies and dividend payments. The first part of the analysis used path analysis and locally weighted estimate scatterplot smoothing (LOWESS) to determine the factors that influence dividends payments. The analysis showed that the stock price and cash balance were stronger determinants for dividends than income.

The latter analysis looked at a dividend capture strategy, which is designed to buy and hold the stock for just a few days to capture the dividend. The data showed that the price of the stock gets discounted on the ex-dividend date (the day after the last day to qualify for dividends), but the discount was not significant between the five-day and fifteen-day intervals. A recurrent neural network was also used to predict the forward five-day average for Verizon’s stock price. The model was relatively accurate for a level one analysis that only looks at the price without considering volume and liquidity.

Indexing (document details)
Advisor: Korosteleva, Olga
Commitee: Suaray, Kagba, Ziemer, William
School: California State University, Long Beach
Department: Mathematics and Statistics
School Location: United States -- California
Source: MAI 81/4(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Statistics, Finance
Keywords: Buybacks, Dividends, Equities, Ponzi, Return, Stocks
Publication Number: 22587243
ISBN: 9781687913791
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest