| The soybean industry was the most important part of the agricultural economy in Heilongjiang Province and was the pillar industry of Heilongjiang Province. The soybean production factor anal-ysis model and forecasting system was of great significance to the sound development of the soyb-ean industry of Heilongjiang Province.This paper based on the soybean acreage and soybean yields which are two basic influence factors of soybean total yields.We established a structural framework about soybean production factors,on this basis,We analyzed soybean production factors and predicted soybean production factors,soybean total yields and soybean yield. The main finding-s and conclusions include the following aspects:We started the analysis from the soybean production status of Heilongjiang province in China, soybean production of Heilongjiang province has been far ahead of other provinces, autonomous regions and municipalities by observing over the years soybean total output datas of the provinces and it played a decisive role in China’s soybean production; we calculated the revealed comparative advantage of china’s provinces, municipalities, district with revealed comparative advantage index; Lateral view,the revealed comparative advantage index of Heilongjiang province was the highest in china,but longitudinal observation, revealed comparative advantage had a downward trend in Heilongjiang province. For Northeast, there was a diffusion trend form Heilongjiang to Inner Mongolia about revealed comparative advantage. According to the changes of soybean total output, soybean total output can be divided into a period of rapid growth and fluctuations in growth period in Heilongjiang province.There were mang characteristics of soybean production, For example, the proportion of soybean total yield was declining in grain output of Heilongjiang, yields and Freque-nt fluctuations in soybean acreage.On the basis of analysis of the status of soybean production in Heilongjiang Province, we established a structural framework about soybean production factors and established a C-D produc-tion function model, stepwise regression production function model and principal component regression production function model.then we analyzed the factors of the soybean total output.The results show that soybean acreage, the hazard rate and time trend were significant impact on soybe-an total output, laborinput fertilizer input as well as effective irrigation area was not significant; Then we established output value of the supply response model between soybean acreage and the output value of soybean and corn,and we also established the profit supply response model between soybean acreage and the profit of soybean and corn. Whether it is the output value of the supply response model or profit supply responsemodel,the results show that comparative advantage of the soybeans directly limit the expansion of soybean acreage and soybean output value, corn output value as well as profit of soybean and corn twas significant impact on soybean acreage, but farmers on the output value of the reaction model was more sensitive than profit reaction model, may be due to farmers’ cost of labor input response was not sensitive. We combine soybean output value of the supply response model and principal component regression production function model and builded the influencing factors chart to analyze the relationship between soybean total yield an-d its factors.We established the stochastic frontier production function model and analyzed the influencing factors of soybean yields from the point of view of technical efficiency. The results show that the contribution of input of amount of labor per mu on soybean yield was becoming weaken; Seed inputs contributed greatly,and increasing seed inputs could improve soybean yield; The contribution of input of chemical fertilizer on soybean yield was becoming enhance; Except the in-vestment in fertilizer and seed,the contribution of other direct material inputs and indirect material inputs were not significant on soybean yield.The average technical efficiency of soybean producti-on was0.9265in Heilongjiang Province,we could improve the technical efficiency in future. Drought rate and the flood rate is a negative impact on soybean production efficiency in Heilongji-ang Province,the impact of the drought rate greater than impact of the flood rate on soybean production efficiency. Single increase the proportion of soybean acreage in the crop area could not effectively improve the yield of soybean,it even reduced the technical efficiency of soybean production. Finally, we combined soybean output value of the dynamic supply response modeland the stochastic frontier production function model and builded the influencing factors chart to analyze the relationship between soybean total yield and its factors.On the basis of the analysis of influencing factors of soybean production,we proposed a time series prediction algorithm based on phase space reconstruction and predicted impact factors of the soybean, soybean output value and corn output value.According to the strong randomness features of the impact factors of soybean yield, we forecasted the drought rate and the flood rate using a Monte Carlo algorithmon.On the basis of the predictive factors, we established the production function prediction model and support vector regression machines prediction model from production function and non-linear regression point of view.Then we established combination forecasting model based on the above two predictive models. We predicted soybean total yield and yield until2015.The research results could offer basis for dominating responsible institution of agriculture in implementing the correct macro-direction and managenment,so that take measures to adjust and control in right timely and appropriately,accomplished the goal of effectively ensuring the goodness of Heilongjiang Province. Moreover,the research methods could offer the theoy to the researchers who are in relevant fields to consult,further to apply it to other domains. |
PDF Full Text Request