Analysis And Forecast On Factors Of Price Fluctuation About Heilongjiang Maize

In 2017,China's maize output reached 259 million tons,ranking first in domestic crop output.Among them,the maize output of Heilongjiang province has been growing at an average rate of 10% per year since 2005,and the annual output reached 31,274,000 tons in 2017,accounting for 12% of the national output in the same year,making it the first maize planting province in China.As a big province of traditional corn planting,the price fluctuation of corn will affect the related interests of corn growers in Heilongjiang province to a large extent.By studying the law of corn price fluctuation in Heilongjiang province,finding out the main influencing factors,which can provide scientific and effective basis for corn growers,corn processing enterprises and the government to make relevant decisions,and timely adjust the production scale to avoid major losses.How to find the relevant influencing factors accurately,build a reasonable corn price model in Heilongjiang province,and effectively predict the price fluctuation trend have become the practical problems that relevant researchers must face,and also become the focus of this paper.The following paper will search for price-related influencing factors from the perspectives of policy,supply and demand,and build a corn price model in Heilongjiang province after factor selection and conduct precision test.After the accuracy is confirmed,the model is used to make short-term prediction and make Suggestions on the analysis results.First of all,as a commodity,the price of corn will be greatly affected by the policy.In order to study the fluctuation of the price of corn caused by policy changes in a more specific way,this paper adopted the policy analysis matrix to carry out relevant calculation and analysis of the corn policy.Through the analysis of the deviation of policy analysis matrix,the relevant data such as profits,could be divided into two different policy in Heilongjiang province corn prices protection period: ? period for strong policy to protect corn prices in 2005-2014 period,? period for weak policy to protect corn prices in 2015-2017.The segmented study can ensure that the influence of other factors on the price fluctuation of corn is not covered up due to the interference of policy factors on the premise of fully considering the influence of policy.Then,the correlative variables of corn price models in different time periods were selected.In this paper,correlation test and random forest importance ranking are combined to ensure the accuracy of variable screening.After the screening,logarithmic processing was performed on the selected related variables to eliminate the influence of heteroscedasticity on the model.Meanwhile,root stability test was conducted on the data to make necessary preparations for the establishment of corn price model.After the preparation,the paper chooses vector auto regression model(VAR)to construct the corn price model in different periods.By testing the accuracy of the price model,the accuracy of the corn price model established according to different policy protection intensity is more than 85%,which has the value of practical research.In order to further study the fluctuation law,the variance decomposition of corn price models in different time periods was carried out in this paper.According to the analysis results,the main influencing factors were selected for pulse fluctuation analysis,and their influence on corn price in different price models was explored.The research results show that the current price of corn,corn futures prices is the first ? period fluctuations in the price of corn in the main influencing factors,the contribution degree of fluctuations in the price of corn was 45% and 22%,respectively.The current price of corn,corn futures price and import price of corn for the first ? section of the main factors that influence the fluctuations in the price of corn,the contribution degree of fluctuations in the price of corn respectively reached 50%,20% and 15%.Finally,this paper selects the weak policy protection corn price model to carry on the short-term price forecast.By substituting the data,it can be seen that the error between the predicted value and the actual value obtained by the price model is less than 10%.Therefore,it is of practical significance to judge the prediction data,and the corn price prediction model can provide an effective reference for the price prediction.