基于混沌理论的建设项目环保投资估算研究

采用混沌时间序列分析技术对我国高速公路环保投资做了较为全面的研究，分析了1997年—2006年我国部分高速公路环保投资的数据。首先对数据结构进行定性分析，通过频数分布图，发现环保投资数据与正态分布之间存在着差异。功率谱分析、主分量分析则清楚表明序列具有非线性性质，主分量分析还表明序列具有混沌迹象。在定性分析的基础上，通过计算序列的有关非线性特征值对其混沌性质进行进一步确认。采用G—P算法计算序列的相关维，采用小数据量法计算最大Lyapunov指数。计算得到的最大Lyapunov指数为正数，说明环保投资序列处于混沌状态，可以使用混沌时间序列方法对建设项目可行性研究阶段环保投资进行估算。

Research on Environmental Cost Estimation of Highway Construction Projects Based on Chaotic Theory

This paper adopts Chaos time series to forecast environmental cost of highway construction projects in China. The data of environmental cost of some highway projects from 1997 to 2006 are analyzed. First of all, the data structure is analyzed by qualitative analysis, such as frequency, power spectral analysis and principal content analysis （PCA）. Through the frequency distribution statistics, the differences have been found between environmental investment data and normally distributed random data. Nonlinear character of the time series is determined by the power spectral analysis, and the chaos character of the time series is easily illustrated by the PCA. Moreover, significantly chaotic dynamics can be further confirmed in terms of calculating the non- linear property value of the time series. G-P algorithm is employed to determine correlation dimensions, and small data set method is employed to determine maximum Lyapunov exponents. The maximum Lyapunov exponent of the time series is positive number, which means that the environmental investment time series is in chaotic state and chaotic time series can be used to predict environmental investment of construction projects at feasibility research phase.