云南城市化水平预测分析——基于ARIMA模型.doc

云南城市化水平预测分析——基于ARIMA模型 摘 要 时间序列分析是一种应用广泛的数量分析方法，它主要用于描述和探索事物随时间发生变化的数量规律性。时间序列的分析模型主要有ARMA 模型和ARIMA 模型。近几年，国内很多专家学者对此及相关领域进行了研究，并且建立了相应的时间序列分析模型。本文根据云南统计年鉴1987~2010 年的非农人口和总人口人口数据，建立了基于ARIMA模型，实验结果表明该模型的预测效果较好。通过该模型对1993年和2006年云南城镇人口比重进行预测，得出预测结果的所有预测误差的绝对值都小于1%。而且，用该模型预测未来2011年的云南城市化水平，从对2011年的预测结果36.95%来看，预测结果与云南省政府提出的在2011年前后云南城镇化水平将接近36.8%的预期目标基本一致。 关键词：城市化；ARIMA模型；预测 Yunnan province urbanization level forecast analysis, based on the ARIMA model Abstract:The number of time series analysis is a widely used analytical method, it is mainly used to describe and explore things change over time the number of regularity. Time series analysis of models mainly include ARMA model and ARIMA model. In recent years, many domestic experts and scholars and related fields have been researched, and the corresponding time sequence analysis model is established. According to the statistical yearbook from 1987 to 2010 in yunnan population data of non-agricultural population and the population, based on ARIMA model is established, the experimental results show that the model has good prediction effect. Of 1993 and 2006 through the model to predict the urban population proportion in yunnan, the forecasting results of all the absolute value of the prediction error is less than 1%. And use the model to predict the future level of urbanization in yunnan province in 2011, from the forecasting results of 2011 36.95%, predicting results and put forward by the government of yunnan province in 2011, before and after the yunnan province urbanization level will be close to 36.8% of the expected goal. Key words: urbanization;ARIMA model; prediction 目录 1 引 言 1 1.1 选题背景及选题意义 1 1.1.1 选题背景 1 1.1.2 选题意义 2 2 文献综述 3 2.1工业化与城市化关系的研究 3 2.2 全球化视角下对中国城市化动力的研究 5 2.3 对国内外研究状况的评述 7 3 云南城市化水平的演变过程 7 3.1 城市化水平的趋势 7 3.2 城市化水平描述性分析 8 4 云南城市化水平模型的构建 9 4.1 原序列的平稳性判定 9 4.2 一阶差分后序列平稳性的判定 10 4.3 ARIMA模型的建立 10 4.4 ARIMA模型的检验 12 4.5 模型的预测 13 5 提升云南城市化水平的策略 14 6 结论 16 6.1 主要发现 16 6.2 启示和努力方向 16 参考文献 17