A hybrid forecasting method for quantitative investments based on GA-VMD and SSA-DELM optimization

In the modern financial market, quantitative investment is a more advanced method, which has been widely used in capital market with characteristics of simplicity, efficiency, and plasticity. Compared other countries, China started late field, strategies are not perfect enough to be studied depth. To study in-depth, this paper proposes GA-VMD-SSA-DELM model. By building an extreme learning machine (DELM) model optimized by sparrow algorithm (SSA), simultaneously using variational modal decomposition (VMD) genetic (GA) for data noise reduction, constructed, portfolio strategy formed. The method first performs correlation tests on sources, uses daily frequency trading indicators Guizhou Maotai, Zhifei Bio, Yangtze River Power years 2018-2022 as database, SPSS conduct Pearson coefficient analysis establish between data. Then GA-VMD was reduction. Finally, SSA-DELM model, construct portfolio, draw relevant conclusions.