Research On Option Price Prediction And Option Strategy Based On Deep Learning

As one of the cornerstones of financial derivatives,options are important in risk management in the financial sector and have excellent arbitrage and hedging properties.Traditional no-arbitrage option pricing methods,including the Black-Scholes option pricing model,the Merton model,and the Heston model,all use stochastic processes to fit the option pricing process under strict assumptions.However,due to the difference between their assumptions and the real market environment,these methods cannot accurately model the option price movements in the real market.Therefore,this paper attempts to solve the option pricing problem using deep learning algorithms from a data-driven model.This paper selects data of SSE 50 ETF options and investigates the feasibility of using time series data processing methods and deep learning models for European-style option pricing based on Black-Scholes option pricing theory.Two option price prediction models are established and the option prices are predicted using these two models respectively,and the three error indicators,MSE,MAE and R-squared,are used to evaluate the prediction accuracy of different models.The empirical results show that the forecasting accuracy of the LSTM model has a significant advantage in predicting the SSE 50 ETF option price.Next,this paper investigates the trading data of SSE 50 ETF options in the first half of 2022 using the vertical arbitrage strategy and the cross-period arbitrage strategy,respectively.The empirical results show that transaction costs and liquidity impose constraints on arbitrage opportunities and arbitrage returns,but there are still many reverse arbitrage opportunities in the parity arbitrage process.Overall,there are still more arbitrage opportunities in China’s SSE 50 ETF options market as a whole.This proves that in the process of exploring the option arbitrage strategy in the Chinese stock market,it can achieve the purpose of reasonable allocation of capital and can correct the price deviation of the market to a certain extent,which has important practical significance.