Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Bachelor's Thesis |
| Scope | Discipline-based scholarship |
| Title | Is the value premium in influenced by investor sentiment? - An empirical investigation for the German market |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Economics, Business Administration and Information Technology |
| Date | 2011 |
| Zusammenfassung | Within the nancial literature, the question what actually drives asset prices on nancial markets is still controversially discussed. On the one side, there are researchers postulating that asset prices re ect all publicly available information a ecting the assets' expected future cash ows. In sum, behavioral aspects like investors' sentiment have no in uence on asset prices because the arbitrage process ensures that any misvaluation caused by noisy, not-fully-rational investors are instantly eliminated. On the other side, proponents of the behavioral nance approach hint that the overall rational view is descriptively inaccurate which means that in reality there are observable market anomalies like the January e ect or speculation bubbles in general that cannot be entirely justi ed by the rational approach. In some sense, the issue what eventually drives the value premium anomaly is exemplary for the two distinct approaches in nance. Fama and French (1998), for example, not only prove that there is an international value premium, i.e. stocks classi ed by high book-to-market equity ratios earn higher average returns than stocks with low book-to- market equity ratios, but also suggest that a two-factor model, including a risk factor, captures the value premium in international returns. By contrast, Lakonishok, Shleifer and Vishny (1994) state that the value premium is mainly attributable to investors' overreaction. This prevailing controversy triggers our interest of research, prompting us to do research on the issue whether the value premium, taken by itself, is in uenced by investor sentiment. Intuitively, as long as humans actively participate in nancial markets we cannot categorically deny that behavioral factors have an in uence on asset prices not least because of psychological elements driving the overall human decision-making process. More speci cally, we choose the German stock market as our benchmark between 1997 and 2010. Thereby, we de ne all shares being listed in the value-weighted CDax performance index of the Deutsche Boerse AG as our universe of German stocks. We obtain adjusted stock price data for all rms listed in the CDax from Thomson Reuters Datastream. In order to adequately discuss our research issue, we proceed as follows: in section 2 we give a review of the existing literature on classical and behavioral nancial economics and discuss, on the basis of Gordon's (1959) growth model, in which way investor sentiment potentially a ects the value premium. In section 3 we begin with our empirical analysis by dealing with the question of quantifying investor sentiment. Strictly I speaking, we gather proxies that all capture some aspects of investor sentiment. In order to avoid measuring non-sentiment-related components, we conduct a principle component analysis. Furthermore we orthogonalize our sentiment proxies to ensure that our results are not predominantly in uenced by economic conditions. Eventually, we refer to the rst principle component as our German Investor Sentiment Index (GISI). We obtain data on sentiment proxies and macroeconomic indicators from Thomson Reuters Datastream, Deutsche Bundesbank, and Deutsche Boerse AG. In section 4 we rst investigate whether there is a value premium in the German market and then perform linear regressions to verify our theory, based on Gordon's (1959) growth model. The results of our empirical investigations, conducted by the statistical analysis software SAS suggest that there is at least no linear in uence of investor sentiment on the German value premium. Nevertheless, we do not categorically deny that there is an in uence of investor sentiment on the German value premium. However, to analyse this issue, further research is necessary. In doing so, this research should not only be capable of capturing over time e ects being responsible for the non-linear relation between the value premium and investor sentiment, but also take the speci c investors composition of the analysed market into consideration. Finally, we point out that data coming from Thomson Reuters Datastream su er from, to some extent, classi cation, coverage, integrity, and survivorship bias. Thus, we emphasise that our empirical results should be taken with caution. |
| Export |
BibTeX
|