The goal of the grant project is the construction and verification of a heterogeneous agent model which will be an extension of the model developed by Brock and Hommes. The new model will include a possibility to change the mood of the investors on the market. This modification will allow changing phases of optimism and pessimism and will enable generation of more realistic financial time series. The outputs of our model will be compared to real market data such as market indices. For an analysis of the results, we intend to use wavelets, which is a relatively new tool for analyzing financial time series. The wavelet analysis is, in some sense, complementary to the existing analyzing techniques, such as the correlation and the spectral analysis, and appears to be a very convenient tool for activity detection on various scales of the price time series. The ambition of the project is to explain some stylized facts in finace with the heterogeneous agent model.