时间序列平稳性和单位位根检验.pptVIP

§81时间序列平稳性和单位根检验 Stationary Time Serial and Unit Root Test 时间序列的平稳性 单整序列 三、单位根检验 ·经典时间序列分析模型: 包括MA、AR、ARMA模型 平稳时间序列模型 分析时间序列自身的变化规律 现代时间序列分析模型: 分析时间序列之间的结构关系 单位根检验、协整检验是核心内容 现代宏观计量经济学的主要内容 、时间序列的平稳性 Stationary Time Series 问题的提出 经典计量经济模型常用到的数据有: 时间序列数据(time- series data); 截面数据( cross- sectional data) 平行面板数据( panel data/time-series cross-section data) 时间序列数据是最常见,也是最常用到的数据。 经典回归分析暗含着一个重要假设:数据是平稳的。 数据非平稳,大样本下的统计推断基础“一致 性”要求——被破怀。 数据非平稳,往往导致出现“虚假回归” ( Spurious Regression)问题。 表现为两个本来没有任何因果关系的变量,却有很高的 相关性。 例如:如果有两列时间序列数据表现出一致的变化趋势 (非平稳的),即使它们没有任何有意义的关系,但进 行回归也可表现出较高的可决系数。 ournal of Econometrics 2(1974)111-120. North-Holland Publishing Company Granger Newbold suggested a rule of thumb when estimating regressions with time series data: if the value of r2 is greater than value of the Durbin-Watsor statistic, then one should suspect a SPURIOUS REGRESSIONS IN ECONOMETRICS C W.. GRANGER and P NEWBOLD UniVersity of Nottingham, Nortingham NG7 2RD, England Received May 1973, revised version received December 1973 It is very common to see reported in applied econometric literature time series regression equations with an apparently high degree of fit, as measured by the coefficient of multiple correlation R or the corrected coefficient R, but with an extremely low value for the Durbin-Watson statistic. We find it very curious that whereas virtually every textbook on econometric methodology contains explicit warnings of the dangers of autocorrelated errors, this phenomenon crops up so frequently in well-respected applied work. Numerous examples could be cited, but doubtless the reader has met sufficient cases to accept our point. It would,for example, be easy to quote published equations for which R=0.99 and the Durbin-Watson statistic(d)is 0.53. The most extreme example we have met is an equation for which R= 0.99 and d=0.093. However, we shall uggest that cases with much less extreme values may well be entirely spurious 2、平稳性的定义 假定某个时间序列是由某一随机过程 ( stochastic process)生成的,即假定时