Portfolio Optimization with Wavelet Analysis and Neural Fuzzy Networks
Ömer Z. Gürsoy and Oktay Taş
The aim of the study is to make short /medium term forecasts about financial asset prices with the help of neural fuzzy networks and wavelet analysis. Studies on the use of wavelet analysis and artificial neural networks for estimating financial assets are available in the literature. In this study, the value estimation for Turkish assets is made with fuzzy neural networks, which are a derivative of artificial neural networks and are used together with fuzzy logic and artificial neural networks. Using the forecasts, buy / sell signals were created and it was decided whether to invest in the relevant asset.
The results obtained in the study reveal that wavelet analysis and fuzzy neural networks have an important potential in the price estimation of financial assets that require in-depth analysis. The wavelet transform technique increases the performance of artificial neural networks and the wavelet neural network method can be used in the price estimation of financial assets such as BIST30, Gold and exchange rate.
Keywords: Portfoliooptimization, wavelet analysis, fuzzy logic, neural networks