Modeling financial markets using mixed minority/majority games

Financial markets are considered to be a system formed due to the interaction between heterogeneous individuals. Many models have tried emulating it and have tried to uncover the working behind it. Minority Game Model is one such model which has tried to emulate it. It is a game consisting of heterogeneous agents who believe that to gain profit one needs to be in the minority. However, it has been proved that the financial market consists of both fundamentalists (i.e. individuals gaining profit by being in the minority) as well as noise traders (individuals gaining profit by following the herd). So, we have used the Mixed Game Model to emulate financial markets which consists of Minority and Majority game players. Although it has been proved that the mixed game model is a suitable model to imitate financial world, we have observed that it still has many limitations like the two groups of agents have same properties and thus they lack in heterogeneity and also that the life of each agent is constant. But in real world, every individual has a unique memory and learning ability and will join and leave the markets as well. To improve on these limitations we have created the model, “Highly Heterogeneous Model” which removes both of these limitations. Also, we show that the new improved game improves the performance of the majority game players by 2.35 % and minority game players by 4.45 %. Apart from this we observed that all the models which have emulated financial market by using Minority Game have concentrated on the combined effect of agents of financial factors like prices, returns and volatilities i.e. they are synchronous. With the availability of high frequency data, its analysis has been continuously gaining importance in recent years. We have thus also studied this behavior of market using the asynchronous form of the game known as the “Asynchronous Mixed Game Model”. We finally also prove that the Highly Heterogeneous Game represents the daily time series and the Asynchronous Mixed Game represents the high frequency time series of real financial world.M.S.Includes bibliographical referencesby Niki Savjan