No 2016/2 - Multi-resolution analysis in economics
The project is dedicated to Discrete Wavelet Transform, which is a linear transformation, like the Discrete Fourier Transform and Discrete Cosine Transform.
In the research numerous properties of wavelets are used, in particular the fact that the Discrete Wavelet Transform gives high degree of freedom to design base functions.
The purpose of this project is the application of multiresolution analysis in different areas of economics, in particular to approximate time series and to construct algorithms for the prediction of time series based on the cognate series with forecasted series in context of the same phenomenon. The project will also examine the issue "edge effects" and their impact on approximation and prediction.
Principal investigator: Monika Hadaś-Dyduch, PhD
Preliminary results of the research in English:
Publications in English: