Dynamic predictor selection and order splitting in a limit order market

Recent empirical research has documented the clustered volatility and fat tails of return distribution in stock markets, and yet the markets are informationally efficient. Certain agentbasedtheoretical models attempt to explain the empirical features in terms of investors' ordersplitting or dynamic switching strategies, both of which are popularly used by actual stock investors. However, little theoretical research has discriminated the behavioral assumptions within a model and compared the impacts of the assumptions on the empirical features. In addition, the research has not simultaneously replicated the return features and empirical features on market microstructure, such as patterns of order choice. This study constructs an artificial limit order market, in which investors split orders into small pieces or interchangeably use fundamental and trend-following predictors over time. We demonstrate that, on the one hand, the market that features strategies with order splitting and dynamic predictor election can independently replicate clustered volatility and fat tails in an informationally efficient market. However, we also show that patterns of order choice do not match those found in certain previous empirical studies in both types of economies. Thus, we conclude that in reality, the two strategies simultaneously work to generate the empirical macroeconomic features, but that investors may also use certain other strategies in actual stock markets. In addition, we demonstrate that the impact of both strategies on the volatility persistence tends to be greater as the number of traders increases in the market; this finding implies that the order-splitting strategy and dynamic predictor selection are more crucial on the empirical phenomena that pertain to larger capital stocks.