| Summary: | 碩士 === 清雲科技大學 === 財務金融所 === 100 === The Forecasting Powers of Gordon’s Dividend Growth Model－Evidence from Taiwan Stock Markets
Student：Meng-Chang Yang Advisors：Prof.Jen-Tsung Hsu
Department of Finance
Ching Yun University
Abstract
The assumptions of the Gordon model are that the dividends paying are continuous and the growth rate of dividends must be positive and constant. In this constraints and assumptions, investors can hardly use it to evaluate their own shares which corporation’s revenue is negative or paying no dividends during their own the shares. In order to solve these problems, we propose the modified Gordon model by using the average dividends yield and the difference of stock returns.
The expected returns of investors are combined with two different parts. One is dividends yield and the other is the growth rate of dividends. The growth rates of dividends are stochastic so it is uncertain for us to use it to evaluate stock values. To avoid such uncertainty, we assume that the growth rates of dividends equals to zero. Also we consider the effects of substituting average dividends yield for the required return rate of the shareholders. Moreover we use the company’s EPS of the first quarter and Earning Variation Rate to forecast the price of the underlying assets after classifying the samples by industries. Then we test the forecasting power of the model by a ratio looks like technical analysis indicator. The modified Gordon model has a better predictive ability in the food, steel and plastics industries, but weaker on the electronics. We look forward to more findings to improve the model.
Keywords: Gordon model, Average Dividends Yield, Growth Rate of Dividends, Earning Variation Rate
|
|---|
