Research On Stock Intelligent Investment Strategy Based On Support Vector Machine Parameter Optimization Algorithm

The changes in the Chinese stock market are inseparable to the dynamics of the market economy of the entire country and have had an important impact on the promotion of the growth of China's national economy.Compared with western advanced capital markets,China's current stock market's main smart investment strategies still have certain deficiencies.In the field of machine learning,the optimization of parameters directly affects the model's core functions and operating performance.The parameter setting mainly depends on personal experience and does not necessarily lead to overall optimality,meanwhile,it is lack of objectivity.These problems have led to errors in the investment of some retail investors and corporate investors.Therefore,it is of great significance to study the optimization of model parameters and the prediction of stocks to guide the reasonable investment of investors.To better predict,this paper further builds a prediction model that combines kernel function and parameter optimization based on the Support Vector Machine model using the grid search method,the genetic algorithm method,and particle swarm optimization under radical basis kernel function,the sigmoid kernel function,polynomial kernel and the liner kernel function to search the parameters of the Support Vector Machine to enhance its practical application.This paper first summarizes the research progress of the Support Vector Machine algorithm and its optimization method,then it elaborates the theoretical basis of the Support Vector Machine,it proposes the idea of the Support Vector Machine prediction model based on parameter optimization and the theoretical basis of the algorithm used.Applying it to the financial market,the short-term stock price of the stock market is forecasted and an empirical intellectual investment strategy for stocks is completed.It provides a new perspective and thought for the analysis of stock speculation strategy.The empirical results show that under the three parameters optimization algorithms,the results are higher than the random prediction's accuracy.This shows that it is effective to optimize the model by optimizing the parameters of the Support Vector Machine model.The Support Vector Machine using the genetic algorithm parameter optimization under the radial basis kernel function shows the best prediction effect,it is the closest to the real value in the stock market prediction.The particle swarm optimization algorithm Support Vector Machine prediction is less effective than the grid search method.In addition,the prediction accuracy of the BP neural network is worse than that of the preparameter tunned Support Vector Machine prediction model.In the end,this paper uses the trained model to plan the stock selection program among stocks.The cumulative yield outperforms the HS 300,the maximum retracement and the Sharpe Ratio are better than pre-tuning model too.In the future,we can consider using a parameter tuning Support Vector Machine model to build a stock intelligence investment model for practical applications.The use of the stock intelligence investment model weakens the individual investor's vulnerability in the capital market,avoids the emergence of some emotional investment,and plays a positive role in promoting the development of quantitative investment in China.The model established in this paper is more economically applicable than the neural network model or other traditional artificial intelligence algorithms and can give investors a certain extent guidance.