On stochastic simulation of stock-exchange

Abstract

A simple Stock Exchange Game Model (SEGM) was introduced in (Mockus, 2003) to simulate the behaviour of several stockholders using fixed buying-selling margins at fixed bank yield. In (Mockus, 2010, 2012; Mockus and Raudys, 2010), the theoretical description of the updated model USEGM was presented and illustrated by the results of experimental investigation obtained by the software developed for the early version SEGM. The new elements of the model is the evaluation of the transaction costs and addition of the minimizer of absolute residuals representing the risk-neutral users. Experimental investigation of the updated USEGM model produced new results. In this paper, we discuss them. The results were compared with real data. They show that the traditional estimators of the minimal prediction errors, such as MSE or MAE, do not necessarily provide maximal average profits. However, contrary to the traditional stock rate prediction models, the main objective of USEGM is not forecasting, but simulation of financial time series that are affected by predictions of the participants.