Research And Development Of Potato Price Analysis And Prediction System

Potato is the fourth largest food crop in China.It is also vegetable,feed and important industrial processing raw material.Develop potato industry vigorously is conducive to ensuring food security and promoting farmer's income.In recent years,potato prices fluctuated frequently,which has a negative impact on the income of potato growers and the healthy and stable development of potato industry.Therefore,based on the analysis of domestic and foreign agricultural product price influencing factors,volatility characteristics,prediction model and analysis and prediction system,this paper studies the fluctuation rule and prediction of potato price,with the key technology of system development,finally designed and developed the potato price analysis and prediction system.The specific research contents are as follows:(1)Analysis of influencing factors and fluctuation characteristics of potato price.The influencing factors of potato price were analyzed by AHP.The price fluctuation of potato is affected by many factors such as previous price,production costs,output,climatic conditions,social conditions and other factors,and the former price has the greatest impact on the potato price.The seasonality,randomness,periodicity and trend of potato price fluctuation in Shandong Province were analyzed by using Census-X12 seasonal adjustment and HP filtering method.Potato prices now generally showed a slight downward trend,the characteristics of the four fluctuation periods are basically the same,the seasonal fluctuations were relatively stable,and the potato price reached the lowest point of the whole year in July every year,and the character of stochastic fluctuation has the most severe fluctuation degree from September 2017 to September 2018.(2)Research on potato price forecast.On the basis of the research on the law of potato price fluctuations,combined with the advantages of traditional time series forecasting methods for processing linear data features and intelligent forecasting methods for processing nonlinear data features,constructed ARIMA,SVM single prediction model and ARIMA-SVM combined prediction model.Using the potato monthly price data in Shandong from January 2011 to June 2019 to forecast,and using RMSE and MAE as the evaluation indicators of the model,the forecast results of a single ARIMA,SVM forecasting model and ARIMA-SVM combined forecasting model were evaluated.The results show that the ARIMA-SVM combined forecasting model has better forecasting effects than the ARIMA and SVM single forecasting model.(3)Research and development of the potato price analysis and prediction system.Based on the analysis of potato price influencing factors,price fluctuation characteristics and price prediction model,combined with the key development technology of agricultural product price analysis and prediction system,the potato price analysis and prediction system was designed and completed.The system include the user information management module,historical data management module,market management module,price fluctuation management module,price forecast management module and policy news management module.The construction of potato price analysis and prediction system can provide accurate data services for many aspects of potato industry,provide more reference information and decision-making services for farmers planning and planting,efficient operation of enterprises,scientific regulation and control of potato industry by government,and finally promote the healthy and sustainable development of potato industry.