Equity Valuation Model and Short-term Market Sentiment: A Quantile Regression Approach

碩士 === 國立成功大學 === 財務金融研究所碩士在職專班 === 102 === The primary purpose of this study is to investigate the relation between equity valuation and short-term market sentiment. This study performs a quantile regression analysis to explore the nonlinear relationship between variables. Our sample covers 1,450...

Full description

Bibliographic Details
Main Authors: Chung-ChingHuang, 黃鐘慶
Other Authors: Hsuan-Chu Lin
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/9mw6j8
Description
Summary:碩士 === 國立成功大學 === 財務金融研究所碩士在職專班 === 102 === The primary purpose of this study is to investigate the relation between equity valuation and short-term market sentiment. This study performs a quantile regression analysis to explore the nonlinear relationship between variables. Our sample covers 1,450 publicly traded firms in Taiwan excluding banking and insurance companies across the period from 2010 Q1 to 2012 Q2, for a total sample of 20,300 panel data. Dependent variables show short-term investor sentiment. Return over the benchmark index, three major institutional investors, sigma, and turnover are first differenced. Main independent variables show residual income return on assets (∆RIOA) and free cash flow return on assets (∆FOA) which all are first differenced. For controlling the size, growth and system characteristics, this study sets up six control variables. Size is log of assets. Growth is log of the duration since first publicly traded and the ratio of price to book value. System characteristics are ∆VIX of TAIEX options, the exchange rate of US dollar to New Taiwan Dollar and ∆M1B. Empirical results show that ∆RIOA has the effect of return-premium and turnover-inspiration. However, it poorly explains the relation between sentiments and institutional investors or sigma. ∆FOA only makes significance on median to high sigma, others are all insignificant. In conclusion, the relation between return, turnover rate and ∆RIOA is more significant than that of ∆FOA. Empirical results also find that the non-linear relation between return on ∆RIOA and turnover rate on ∆RIOA results in underestimation of the OLS coefficient. The results of extreme quantiles ( the left and right tail 10%) find that the market goes more consistent with fundamental valuation in the bull market. In contrast, when the stock turns down, fundamental valuation and other factors such as trading rules would affect the short-term stock price.