RIMA模型在气温预报中的应用--以延安地区为例
he application of ARIMA model in the prediction of temperature--taking Yan\\\\\\\\\\\\\\\'an Region for example
根据延安1961-2000年共40年的七月份月平均气温数据资料分别利用差分自回归移动平均法（ARIMA）对延安2001-2004年七月份月平均气温值做预报。结果表明： ARIMA(1,1,1,)模型用于气温预报时，表现出较强的预报能力，预报值与观测值之间的误差较小，误差绝对值最大仅为2.317℃，相对误差在10%的范围内，可以用于气温预测。但因ARIMA模型自身的不足，它对气温的预报有一定的局限性。为了提高预报的准确度，应当引入或结合其他数学方法，使不同权重的诸多影响气温变化的相关变量参与分析、建模的过程中。从而使ARIMA模型在实际的气温预报时更具实用性。
he paper applies ARIMA model to forecast the average temperature of July from 2001---2004 of YanAN according to the date of YanAn forty years (1961---2000) totally. The result shows that ARIMA(1,1,1) model performed well in temperature prediction. Errors between predicted and observed temperature values are minor, and they are in the acceptable range. The max absolute error is just only 2.317℃, furthermore, relative error is less than 10%. So, the ARIMA model can be used for temperature prediction. But because of ARIMA model\\\\\\\\\\\\\\\'s own shortcomings, its temperature forecast has some limitations. In order to improve forecasting accuracy, other mathematical methods should be introduced or combined with, in addition, it is necessary to make all kinds of different weight-related variables which affect the temperature change involve in the process of analysis and modeling. so that the ARIMA model is more available in the prediction of temperature.
王君兰、李双、郭小燕
大气科学（气象学）
最优气候值法RIMA模型气温预报PS
optimal climate normal methodARIMA modelthe prediction of temperatureDPS
王君兰,李双,郭小燕.RIMA模型在气温预报中的应用--以延安地区为例[EB/OL].(2010-05-19)[2025-08-10].http://www.paper.edu.cn/releasepaper/content/201005-450.点此复制
评论