Natural resistance points at round index values: Theoretical and empirical investigation of the crossing of large round index values and its implications to the development of stock indexes

Georg Boudon

Student thesis: Master thesis

Abstract

The crossing of a popular index through a large round value, such as 6000, receives large attention in the popular media and business press. The media coverage suggests that large round values present resistance or support areas and therefore mark the beginning or end of a trend. Neo-classical finance theory, which relies on the assumption of homo oeconomicus, includes insights such as the Capital Asset Pricing Model, the Arbitrage Pricing Model and the Efficient Market Hypothesis. Those theories are incompatible with the implications of resistance points around large index values. According to those theories, every cycle in stock market development is self destroying and as a consequence, no point or pattern in chart development should have any predictive qualities. Since the excessive media coverage of passages of influential indexes through round values cannot be explained by means of neoclassical finance theory, as it is taught at most universities, the question remains if there is any basis to the large attention it receives. The practice of technical chart analysis, opposed to neo-classical theory, entails that certain patterns in chart development do indeed have predictive qualities and can be profitably exploited. Resistance and support areas, even though not necessarily at round values are very central to technical analysis. The main criticism for technical chart analysis is its lack of a sound theory. That criticism cannot be dismissed but behavioral finance and cognitive science suggest that the potential mistakes of market agents are highly correlated. Neo-classical theory assumes that mistakes will be neutralized through arbitrage but other cognitive science postulate that certain mistakes are caused by behavioral predispositions and can hardly be avoided. The patterns used in technical chart analysis could be a reflection of such biases. Empirical tests of the predictive abilities of technical trading strategies are usually not successful if data snooping bias is taken in to account. An investigation, particularly directed to resistance points at large round values showed similar results: There is an abundance of heuristics, such as availability bias and anchoring effects, suggesting that a particular significance of large round values for the index development is likely. Nevertheless, the significance of such values could not be proven in an empirical investigation. Behavioral Finance, as well as other cognitive sciences strongly suggest that large round index values influence traders more than other random values. However, if such an influence exists it does not seem systematic enough to prove it empirically or exploit it profitably

EducationsMSc in Applied Economics and Finance, (Graduate Programme) Final Thesis
LanguageEnglish
Publication date2010
Number of pages74