Contrary Opinion Phenomena in an Artificial Stock Market

This paper presents an artificial market model to clarify the definition of the contrary opinion phenomena. Our research target is to show the contrary opinion, which is based on market participants’ experiences, statistically. At first, we constructed a more realistic artificial market model and simulated it. Second, we selected ‘highly corresponded forecasts’ among the fitted sample paths. Third, we classified them into three groups: the ones which on a uptrend with the following trend reversal, those which on a uptrend only and otherwise. Finally, we compared between the volume at all the observations and those both before and after at the observation in each group. This is based on a rule of thumb that both acceleration of price and increasing volume can be seen before the turning point in financial markets. The results indicate that the volume at the peak of the price, when a contrary opinion phenomenon occurs, is the largest. While those of other groups were not. This shows the rule of thumb was replicated in our model. Especially, because the contrary opinion phenomena were clarified statistically, these results could be noticeable.