论文部分内容阅读
利用科学的方法提前半年以上对全国粮食产量进行高精度的预测,已成为近年国际边缘科学研究的热点之一;尤其是在我国,农业是国民经济的基础,而粮食又是农业的基础,农业问题尤为重要。因此,粮食产量的预测研究不仅是一项基础科学研究工作,而且它的研究结果直接对国家宏观调控,特别是调整国民经济政策,安排粮食进出口、储存、销售,乃至制定工农业生产计划都有很大的影响。中国科学院粮食产量预测小组在陈锡康研究员的带领下,自1980年开始连续19年对我国粮食产量进行高精度的预测,取得了重大的成果和进展。预测的理论不断改进,预测精度和预测提前期均居国际领先水平。粮食产量受很多因素制约,如何能提前准确地预报其产量,各国方法各异。国外目前主要采用气象预测法、遥感技术法、统计动力学模拟预测法来进行粮食产量预测。这些方法的预测提前期通常在2个月之内,预测误差为5—10%。而中国科学院粮食产量预测小组经长期研究,提出了以投入占用产出分析和变系数预测方程为核心的系统综合因素预测法,这是一种完全创新的理论和方法,它预测的提前期可在6个月以上,19年来预测的产量与国家统计局来年公布的实际粮食产量相比较,数值非常接近,平均误差只有1.6%,这一理论的科学性和方法的精确性已充分被实践所证实。在今年8月召开的第十五届国际运筹学会联合会学术大会上,中国科学院系统科学研究所陈锡康研究员和潘晓明、杨翠红博士的论文“中国粮食产量预测”荣获“运筹学进展奖”一等奖。——编者
Using scientific methods to predict the grain output in the country more than six months in advance has become one of the hot topics in international and regional scientific research in recent years. Especially in our country, agriculture is the foundation of the national economy, and grain is the basis of agriculture. Agriculture The problem is especially important. Therefore, the forecast study of grain yield is not only a basic scientific research work, but also its research results directly impact on the macroeconomic regulation and control of the country, especially the adjustment of national economic policies, the arrangement of grain import and export, storage and sale, and even the formulation of industrial and agricultural production plans Have a great impact. Under the leadership of researcher Chen Xikang, the prediction team of grain yield of the Chinese Academy of Sciences has made the forecast of grain output in China for 19 consecutive years with high precision since 1980, and made great achievements and progress. The theory of prediction is continuously improved, and the prediction accuracy and prediction lead times are among the highest in the world. Food production is constrained by many factors. How can we predict its output in advance and accurately? Different countries have different approaches. At present, the forecasted forecast of grain yield is mainly based on meteorological forecasting, remote sensing and statistic dynamic simulation forecasting. The forecast lead times for these methods are usually within 2 months with a prediction error of 5-10%. However, based on the long-term research, the forecasting system of grain yield of Chinese Academy of Sciences puts forward the forecasting method of comprehensive factors based on the input-occupancy output analysis and the variable coefficient forecasting equation, which is a completely innovative theory and method. The forecasting lead time In six months or more, the predicted production over the past 19 years is very close to the actual food production released by the National Bureau of Statistics in the coming year, with an average error of only 1.6%. The scientific nature of the theory and the accuracy of the method have been well established Confirmed. In August this year, held in the fifteenth session of the International Federation of Academic Research Conference, CAS Institute of Systems Science researcher Chen Xi Kang and Pan Xiaoming, Dr. Yang Cuihong thesis “China's grain output forecast ” won the "Operation Research Progress Award First prize. --editor