Agent-based modeling

Banks are the financial institutions that collect assets from various sources and allocate them to the sectors that require liquidity. Therefore, banks are an inherent element in the system of every country. As private banks enter financial markets, the demand for diverse banking services increases dramatically. Banks seek to use various techniques to improve their performance in attracting customers to increase their market share and profitability. In this regard, assessing the performance of banks is of utmost importance and has become a major activity of bank managers. With the constant changes in the modern world and incessant attempts of competitors to increase their market share by gaining competitive advantage, special attention should be paid to ambidexterity as a key strategy to increase competitive advantage and achieve high performance in dynamic business environments. The present study aimed to identify the ambidextrous factors affecting the performance of banks and present a model to assess the performance of an ambidextrous bank using an agent-based modeling approach. The main objective of the research is to achieve an applied model for managing the performance of the banking industry. The simulation model is processed using the agent-based modeling approach in AnyLogic software environment.

Expansion of urban area is a well-known phenomenon in developing countries with population growth and the migration from villages to cities being two major factors. Those factors reduce the influence of efforts to limit the cities boundaries. Thus, spatial planners always look for the models that simulate the expansion of urban land-uses, and enable them to prevent unbalanced expansions of cities, and guide the developments to the desired areas. Several models have been developed and evaluated for simulating urban land-use expansions. Although these models are numerous, most of them have focused to simulate urban land-use expansions in sub-urban areas. The regional models that cover wider area are equally important. In this study, a new agent-based model has been developed and implemented to simulate urban land-use expansion in Qazvin and Alborz regions of Qazvin province, which cover 1620 square kilometres. In this model, land-use developers have been treated as computer agents that move in the landscape explicitly, and assess the state of parcels for development. The environment of the model is raster. The agents are categorized based on two scenarios. In the first scenario, all agents are of similar category and in the second scenario the agents are divided into five categories with different objectives. Then, the results of the two scenarios are compared. Due to the spatial essence of the problem, Geographical Information Systems (GIS) were used to prepare the environment of agents’ movement and search, and to aggregate and analyze the results. To evaluate the model, data of year 2005 were used as the input and data of year 2010 were used for checking the results. By calibrating the parameters, the most desired configuration of the model was found in the second scenario, since the results were close to the reality as the Kappa index raised up to 78.17 percent. These results showed that the precision of the model to simulate land-use developments are of considerable quality. Thus, the model is able to detect the area that faced rapid urban expansions. Moreover, a comparison between the results of the two scenarios revealed that dividing the agents into categories with different aims and parameters will improve the outcome of the model. However, it is vitally important to determine the number of the agents in each category as well as their parameters precisely.

Knowing the current public opinion and predicting its trend using opinion formation models is very applicable. The social impact model of opinion formation is a discrete binary opinion model. It describes how interactions among individuals and sharing their opinions about a specific topic in a social network affect the dynamics of their opinions and form the opinion of society. The society could be an online social network. In this research, we considered the effect of segregation on opinion formation. Segregation is a phenomenon that happens due to homophily and is measured based upon network topology. Homophily is the tendency of individuals to interact with others who share similar traits. We used scale-free networks to model interactions between individuals. The social impact model includes a noise parameter, which is the stochastic part of the model, dealing with the inexplicable behavior of individuals and the effects of other influentials, e.g., mass media. Since this noise is a white noise with no bias toward any possible opinion, for simplicity, we assumed a noise-free social impact model, which is valid in equilibrium analysis we considered. The results reveal that with the same attributes for the individuals, the more segregated opinion group dominates the less segregated opinion group on average. Therefore, with the same population size and individual characteristics of both opinion groups, segregation is an overall influential factor for opinion formation. A more segregated opinion group attracts some individuals from the other group and becomes the majority opinion group of society in equilibrium.

Although theoretical and empirical literature regarding the stylized facts shows evidence of their correlations to herding behavior in financial markets, the causes of such phenomena are still unknown. Using an agent-based model strengthened by the competition co-evolution algorithm (STGP) technique, this study provides laboratory evidence on capital market dynamics and analyses the behavioral foundations of stylized facts such as fat tails, leverage effects, and volatility clustering. The simulated stock markets consist of two groups; the “Best agents”, which are a small portion of artificial agents, and the “Residual agents”, which are the main group of artificial agents. The best performance in terms of breeding fitness returns is the main feature of the “Best agents”. More, the size of the “Best Agents” group is specified as 2.5%, 5%, 10% &20% of the total population size. An agent-based model consists of two portions, a two thousand population of trader agents that each has its decision-making strategy, and a virtual market that creates the trading strategies. Then the model evolved step by step using a feed with real quotes of the financial instruments by Adaptive Modeler. A training period is considered 2500 bars (started in November 2003), and the test period started in December 2013. The observation shows that the herding behavior in the price series created by the “Residual agents” is less than the “Best agents” series. Therefore, the greater diversity of trade strategies as the genetic differences of artificial agents leads to less herding. The observations exhibit that the volatility clustering, leverage effects, and nonlinear dependence are more likely to experience in the price series generated by “Best gents”. Furthermore, observations indicate that if the population is well diversified in terms of trading strategies, the efficiency of the market increases.