Research On Forecasting And Quantitative Trading Strategy Of Hog Futures Price

As an important component of the national economy,the hog industry is a source of livelihood for many farmers and is also involved in food security protection.The current development of hog industry has problems such as strong cyclicality and high degree of price volatility,which cause impacts on related industries and residents’ lives.In order to further develop the hog industry,China launched hog futures in 2021 at the Grand Mercantile Exchange.However,at this stage,compared with other agricultural futures,the price of hog futures is not only more volatile,but also more seasonal,which increases the investment risk faced by investors and makes them less willing to invest,resulting in the futures not being able to play their role fully.Therefore,it is important to study the content of hog futures price forecast for investors’ scientific investment and related enterprises’ reasonable production planning to maintain the stability of hog market.Literature combing reveals that literature on hog futures mostly focuses on the feasibility and significance of listing,with less research on the existing characteristics of hog futures;while literature on the construction of price forecasting models mostly focuses on price forecasting of some futures species through advanced artificial intelligence algorithms,but is not quite suitable for the current characteristics of short listing time and small data volume of hog futures in China,which makes the model system of hog futures price forecasting for China The relevant model system of futures price prediction is relatively thin.At the same time,there is a lack of exploration in the field of quantitative investment for China’s hog futures.Based on this,this paper systematically analyzes the spot and futures markets of China’s hogs and analyzes the reasons why hog futures are not fully effective.The GM(1,1)model and LSTM model are selected to forecast the price of China’s hog futures,and the GM(1,1)-LSTM composite model is explored to prove that the composite model can accurately forecast China’s hog futures through cross-sectional comparison,and the quantitative trading timing strategy is constructed.The main research findings are as follows:(1)The analysis of China’s hog spot market and futures market shows that the role of futures market has not been fully played at this stage.Although futures have played a role in improving the quality of industry and adjusting the production of hog industry,the overall market trading volume is low,which makes the role of hedging and price guidance not fully demonstrated.The main reason for this phenomenon is that investors are not mature enough,do not have the right knowledge about hog futures prices and are not willing to enter the market.(2)GM(1,1)-LSTM model has a high accuracy in predicting hog futures prices.Through the analysis,it is obtained that China’s hog futures are characterized by little data and outstanding periodicity,so the GM(1,1)model,which is less dependent on historical data,and the LSTM model,which has good learning ability for historical data,are selected and combined to get the composite model GM(1,1)-LSTM model,and after the comparison of error values,it is found that the composite model has the best performance in terms of RMSE,MAE and other indicators.The model can have some reference value for this type of financial market forecasting with small data volume and strong cyclicality.(3)The quantitative trading timing strategy was constructed based on the values predicted by the composite model,and achieved better risk-return performance in the backtest,which has practical quantitative investment value,indicating that the model can provide some methods and ideas for quantitative trading investors in the financial market.