Advanced

Beating the Benchmark - an Active Trading Strategy Based on Contrarian Trends and Jump Detection

Svensson, Linus and Fernström, Gustav (2006)
Department of Economics
Abstract
Extreme price changes are identified in the stock market and related to jumps in the price process. After such extreme events, a reverting (also called contrarian) trend is examined. There is evidence that the initial price change is often an overreaction, caused by the psychology of market participants, and that the subsequent reversal corrects for the temporary inefficiencies in the market. Confirming the empirical findings by other studies, it is found that larger extreme price changes are followed by a stronger reversal effect. From a statistical analysis of the extreme events, a systematic trading strategy is formed. After an extreme price change, losing stocks are bought with the hopes of following a reverting trend. In the long run,... (More)
Extreme price changes are identified in the stock market and related to jumps in the price process. After such extreme events, a reverting (also called contrarian) trend is examined. There is evidence that the initial price change is often an overreaction, caused by the psychology of market participants, and that the subsequent reversal corrects for the temporary inefficiencies in the market. Confirming the empirical findings by other studies, it is found that larger extreme price changes are followed by a stronger reversal effect. From a statistical analysis of the extreme events, a systematic trading strategy is formed. After an extreme price change, losing stocks are bought with the hopes of following a reverting trend. In the long run, probability is assumed to be in favor of the strategy, generating significant positive returns. Returns from trading are evaluated for both an estimation and validation period. Trading performance is excellent during the estimation period, beating the defined benchmark. During validation, the active trading also generates promising results, outperforming the benchmark. It is however suspected that the nature of the extreme events change depending on current market conditions. During strong positive market trends, the reverting trends are likely to be more significant. (Less)
Please use this url to cite or link to this publication:
@misc{1334452,
  abstract     = {Extreme price changes are identified in the stock market and related to jumps in the price process. After such extreme events, a reverting (also called contrarian) trend is examined. There is evidence that the initial price change is often an overreaction, caused by the psychology of market participants, and that the subsequent reversal corrects for the temporary inefficiencies in the market. Confirming the empirical findings by other studies, it is found that larger extreme price changes are followed by a stronger reversal effect. From a statistical analysis of the extreme events, a systematic trading strategy is formed. After an extreme price change, losing stocks are bought with the hopes of following a reverting trend. In the long run, probability is assumed to be in favor of the strategy, generating significant positive returns. Returns from trading are evaluated for both an estimation and validation period. Trading performance is excellent during the estimation period, beating the defined benchmark. During validation, the active trading also generates promising results, outperforming the benchmark. It is however suspected that the nature of the extreme events change depending on current market conditions. During strong positive market trends, the reverting trends are likely to be more significant.},
  author       = {Svensson, Linus and Fernström, Gustav},
  keyword      = {Contrarian trading,Extreme price changes,Jump detection,Black-Litterman,Economics, econometrics, economic theory, economic systems, economic policy,Nationalekonomi, ekonometri, ekonomisk teori, ekonomiska system, ekonomisk politik},
  language     = {eng},
  note         = {Student Paper},
  title        = {Beating the Benchmark - an Active Trading Strategy Based on Contrarian Trends and Jump Detection},
  year         = {2006},
}