Research On Predication Of Internet Financial Returns Based On The PSO-LSSVR

In recent years, the internet has developed rapidly, and a series of internet technologies and applications, such as Cloud Computing, Internet of Things, Fog Computing, AlphaGo, etc. emerge accordingly, impact the traditional industries strongly. Internet banking is in Internet technology innovation and the traditional financial sector integration background, in particular, its’ rate has a profound effect to guide financial investment, promote China’s interest rate in market and develop the inclusive financial system, therefore financial forecast yields are particularly important.In this paper, we used Back Propagation neural network, Wavelet neural networks, Support vector regression(SVR) and Particle swarm optimization of least squares support vector regression(PSO-LSSVR) to study the seven-year rate of return of Yu’ebao, and then use the test data to test the effect of the simulation model, the results showed that: SVR and PSO-LSSVR predict better than neural network algorithm. Then we choose Jiashi Monetary Fund return data to validate the effectiveness of the four methods, it also showed that: the prediction effect of PSO-LSSVR is better than other methods.It will be a good yield forecast to the trends in Internet finance through empirical analysis, its practical significance in three aspects. First, the IT FIN as an innovative financial platform also has the traditional financial investment risks, the accurately prediction of the rate can guide investors to invest. Second, the IT FIN mechanism is completely market-oriented, studying on IT FIN future trend will help study the process of China’s market-oriented interest rate. Third, the IT FIN as an efficient financial management platform to the masses, we can research the development in practicing the inclusive finance through studying the trend of the IT FIN returns.