Research And Implementation Of Data Quantitative Analysis Strategy Based On Machine Learning

Nowadays,financial institutions rely on analyzing data from both fundamental and technical sources,but these methods have limitations that often lead to decision-making errors.These errors are caused by human weaknesses,which pose a risk to financial activities.In the realm of quantitative investing,predicting stock prices and developing effective trading strategies are paramount.The inherent randomness of the stock market,however,limits the precision of stock price predictions,thus undermining investment performance when these predictions are the sole basis for strategy.Nonetheless,forecasting future prices can offer invaluable insights for strategic models.When predictions are paired with well-designed strategies,they can augment investment returns.This thesis proposes a novel method to improve investment returns,which combines Long Short Term Memory(LSTM)prediction with Evolutionary Operatingweights(EOW)algorithm strategy.The method further improves the accuracy of the prediction model by screening factor pools.And it utilizes multi-layer LSTM to predict future stock prices and combines the predicted results with current market environment data,and then the operation strategy is obtained using the EOW algorithm.This thesis compared with traditional buy-and-hold strategy,as well as models based on turtle strategy,moving average strategy,Q-Learning strategy,and policy gradient strategy.The experimental results show that in different time periods,the method proposed in this thesis can achieve significantly higher returns than the benchmark strategy,and has lower risk.This thesis not only verifies the effectiveness of LSTM network in stock price prediction,but also shows the advantage of EOW in optimizing operation strategy.This thesis provides a novel and effective quantitative analysis tool for financial market investors.