Fractal and chaos theory combination-based financial time series short-term prediction

Abstract

The invention discloses a method for applying a fractal and chaos theory combination to short-term prediction of a price series in the financial field. The method comprises the steps of calculating a Hurst index and a statistic quantity, defined in the specification, of a time series by utilizing an R / S analysis method, and determining a cycle of the time series; working out an embedded dimension and delay time of the time series, and performing phase space reconstruction on the time series to generate sample data; performing normalization processing on the sample data; determining a difference value point according to a sequence of sampling time points; determining a vertical scaling factor by utilizing the Hurst index; determining an iterated function system of each group of samples, and determining a final iterated function system according to a weight; extrapolating a time point, setting a corresponding initial value to be zero, setting a step length for enabling a corresponding value to be transformed, and re-performing iteration; and drawing a curve, and calculating an average error of a difference value result and historical data, wherein an extrapolated value corresponding to a minimum error is a predicted value. The method mainly aims at the short-term prediction of the time series with unremarkable self-similarity and periodicity in the financial field; and a predicted result is more accurate than that of a conventional fractal method.

Description

technical field [0001] The invention relates to a method for short-term forecasting of financial time series by using a method combining fractal and chaos theory, and belongs to the field of financial data mining. Background technique [0002] With the development of financial globalization and liberalization, the competition in the financial industry is becoming more and more fierce. To a certain extent, traditional financial analysis methods can no longer satisfy the dependence and sensitivity analysis of large amounts of data. In recent years, in response to the dynamic, complex, and nonlinear characteristics of financial data, people have introduced nonlinear theory into the financial market in order to more accurately reveal the operating laws of the financial market from these data. [0003] Chaos and fractal are the main theories of nonlinear science. Using nonlinear theory can better analyze dynamic complex and nonlinear big data. At present, relevant research resul...

AI Technical Summary

Problems solved by technology

[0002] With the development of financial globalization and liberalization, the competition in the financial industry is becoming more and more fierce. To a certain extent, traditional financial analysis methods can no longer meet the dependence and sensitivity analysis of large amounts of data.

Images