Machado, J. A. Tenreiro2014-02-062014-02-0620130924-090X1573-269Xhttp://hdl.handle.net/10400.22/3772This paper presents a novel method for the analysis of nonlinear financial and economic systems. The modeling approach integrates the classical concepts of state space representation and time series regression. The analytical and numerical scheme leads to a parameter space representation that constitutes a valid alternative to represent the dynamical behavior. The results reveal that business cycles can be clearly revealed, while the noise effects common in financial indices can elegantly be filtered out of the results.engTime seriesComplex dynamicsFinancial analysisState spaceTrendlinesjournal article10.1007/s11071-013-0965-x