Parameter estimation method of option pricing model based on convolutional neural network in high frequency financial trading

The current option pricing model can more accurately approximate the actual market conditions, but it is difficult to estimate its parameters, and high-frequency financial transaction data analysis requires high real-time performance. All of these make it difficult to estimate the parameters of the option pricing model. Therefore, this article focuses on the parameter estimation problem of option pricing model in the high-frequency financial trading environment, and a two-stage heuristic algorithm based on convolutional neural network is given. The core idea is to train the historical information accumulated on the example, and use this information to guide the solution of the example, to get the parameter estimation result of the option pricing model. Experimental results show that the convolutional neural network stage can well complete the offline learning of the nine main contracts, the two-stage heuristic algorithm based on the convolutional neural network can make full use of historical data information to efficiently solve the problem, and it can be quickly optimized in the optimization stage, achieve convergence and meet the real-time requirements of high-frequency transaction data analysis; after obtaining the option pricing model parameters, the model pricing has a higher consistency with the actual price, and the fitting accuracy is higher. The research in this paper enriches the parameter estimation methods of option pricing models under high-frequency financial transactions, and at the same time expands the application boundaries of heuristic algorithms.