Stock Stock Index Forecast Forecast with Back Propagation Neural Network Optimized by by Genetic Algorithm

Stockindex index forecast is not an easy jobas it it is subject to to influence of various various factors. factors. Since ince 1980s 1980s,, many researchers have have used used Back Propagation Propagation Neural Network ( BPNN BPNN) to to forecast forecast stock price price fluctuations fluctuations. However, there are are some limitations with BPNN. With slow convergent speed and low low learning learning efficiency, efficiency, BP learning algorithm is easy to get in in local minimum and and is is far far from being being perfect perfect in in stock forecasting. The genetic algorithm algorithm is a a sort of self- adaptive optimized optimized optimized optimized search algorithm based on natural selection and and natural inheritance. It can can be applied applied in different areas of of parameter space in the colony generation subrogation toward the optimal direction,, which which the the search could easily find and couldn couldn't get in local minimization minimization minimization minimization. In In view of this, we we