O efeito manada nos fundos de investimento no Brasil: um teste em finanças comportamentais
Using daily flow data on stock, hedge and fixed income funds in Brazil, aggregated by average investor’s size (riches and poors), and by means of a flow direction based methodology, there is strong evidence of herding in a heterogeneous distribution within different groups of investors. The intensit...
| Author: | |
|---|---|
| Format: | master thesis |
| Status: | Published version |
| Publication Date: | 2010 |
| Country: | Brasil |
| Institution: | Fundação Getulio Vargas (FGV) |
| Repository: | Repositório Institucional do FGV (FGV Repositório Digital) |
| Language: | Portuguese |
| OAI Identifier: | oai:repositorio.fgv.br:10438/4921 |
| Online Access: | http://hdl.handle.net/10438/4921 |
| Access Level: | Open access |
| Keyword: | Mercados financeiros e finanças corporativas Finanças comportamentais Efeito manada Administração de empresas Fundos de investimento - Brasil Finanças - Processo decisório Investimentos - Análise |
| Summary: | Using daily flow data on stock, hedge and fixed income funds in Brazil, aggregated by average investor’s size (riches and poors), and by means of a flow direction based methodology, there is strong evidence of herding in a heterogeneous distribution within different groups of investors. The intensity of the herding behavior varies according to the investor size, kind of fund and over time. Furthermore, an heuristic-driven bias was tested: anchoring. It is based on the popular assumption within chart analysts that new highs and lows on the stock market have a meaning over the future prices. In this study, we found evidence that the event of a new high or low on the stock market index Ibovespa has some explain power over the herding measure on all kinds of funds, not only stock funds, and the impact of a new low is stronger than the impact of a new high. Although that effect exists, its explaining power is very low, and this paper’s statistical results suggest that there are other factors, not addressed, that have much more explaining power over the herding behavior of fund investors. In conclusion, this study suggests that three of four behavioral finance assumptions tested are true: the expectations and the information on the financial markets are not homogeneous, and the investors don’t make decisions based purely on maximizing their expected utility, but also imitate other investors. The fourth behavioral finance assumption tested, heuristic-driven bias, wasn’t denied, but has shown very small, if any, relevance on the herding behavior. |
|---|