你重新排一下发啦，这个要运行代码的，麻烦啊。

Perron最近几年在JOE上也连续发了好几篇关于结构断点的文章，我稍后也会把这些文章传到这个帖子上来，希望大家能多多讨论。

Hao, K., “Testing for Structural Change in Cointegrated Regression Models: some
Comparisons and Generalisations,” Econometric Reviews, 1996, (15, 4), 401–429.

Johansen, S., “Statistical Analysis of Cointegration Vectors,” Journal of Economic
Dynamics and Control, 1988, (12), 231–254.

and S. Ouliaris, “Asymptotic Properties of Residual Based Tests for Cointegration,”
Econometrica, 1990, (58), 165–193.

Hao, K., “Testing for Structural Change in Cointegrated Regression Models: some
Comparisons and Generalisations,” Econometric Reviews, 1996, (15, 4), 401–429.

请找下这几个文章。

谁有这几篇文章的，可以传来看看啊。

谁有这篇文章，请发帖

Title: Monte Carlo tests of cointegration with structural breaks
Author(s): Ralf ?stermark, Rune H?glund
Journal: Kybernetes
ISSN: 0368-492X
Year: Dec 2000 Volume: 29 Issue: 9/10 Page: 1284 - 1297
DOI: 10.1108/03684920010346347
Publisher: MCB UP Ltd
Abstract: The power and size of five cointegration tests, the ADF-, Z^a-, ECM-, SW-, and JJ-statistics, are evaluated in some large-scale Monte Carlo simulations, when the underlying system is subjected to regime shifts. Following the suggestion by Gregory and Hansen, selects the minimum value for the shift-corrected statistics evaluated over a set of tentative break points for the regime shifts. The performance of these statistics is compared to the corresponding ordinary statistics in conditions of regime shifts. The results show that no test uniformly outperforms the others in terms of power in the parameter space we have used.

谢谢楼主

Numerical Distribution Functions of
Likelihood Ratio Tests for Cointegration
by
James G. MacKinnon

Summary
This paper employs response surface regressions based on simulation experiments to
calculate asymptotic distribution functions for the Johansen-type likelihood ratio tests
for cointegration. These are carried out in the context of the models recently proposed
by Pesaran, Shin, and Smith (1997) that allow for the possibility of exogenous variables
integrated of order one. The paper calculates critical values that are very much more
accurate than those available previously. The principal contributions of the paper are
a set of data files that contain estimated asymptotic quantiles obtained from response
surface estimation and a computer program for utilizing them. This program, which
is freely available via the Internet, can be used to calculate both asymptotic critical
values and P values.
JEL Classification Number: C22
Keywords: cointegration tests, Johansen tests, vector autoregressions,
exogenous variables, response surfaces, critical values, approximate
P values, simulation

好！谢谢，学习一下！！

******** ADF Test ***********
t-statistic = -4.5918447
AR lag = 6.0000000
break point(ADF) = 0.54687500

******** Phillips Test ********
Zt = -5.8587021
breakpoint(Zt) = 0.50781250
Za = -54.644997
breakpoint(Za) = 0.50781250

这是Bruce E. Hansen程序计算结果，他的文章中画了个图，可以看出断点在哪里。但他的程序没有画图，是根据断点处的t值判断吗？断点怎么会有t值了？邹检验是2个F值啊。请网友指点。

请问在gauss里面.fmt数据格式放哪里？怎么写导入数据。

This program file loads the GAUSS dataset "invest.fmt".
It creates the output file "thresh.out"

*/

load invest;
t = 15;
nt = rows(invest);
n = nt/t;

i = invest[.,1]; @ investment/assets @
q = invest[.,2]; @ Tobin's Q @
c = invest[.,3]; @ cash-flow/assets @
d = invest[.,4]; @ debt/assets @

Tests for Parameter Instability and Structural Change With Unknown Change
Point
Donald W. K. Andrews

1993年的虽然有点老，但是里面的估计方法很新，perron1998和2007年的两篇论文都参考此文献。

利用了部分gmm估计求未知突变点的wald，LM和LR统计量，推导了零假设的渐进分布，并进行了模拟。

零假设没有结构突变，备择假设有结构突变

文章很长，数学太多，里面的推导我没有怎么看，只是理解了一些主要内容，文章内容还是不错的。

传3篇Perron2003年之后发在JOE上的文章，

Estimating restricted structural change models(Perron 2006)

Structural breaks with deterministic and stochastic trends(Perron2005)

GLS detrending, efficient unit root tests and structural change(Perron2003)

对LM检验方法，请问xuelida有研究吗？国内好像是用这种方法的多一些。

放假都没有人了？

有人啊,搞个程序大家琢磨琢磨?

1995An LM Test for a Unit Root in the Presence of a Structural Change.pdf

2004Minimum LM Unit Root Test with One Structural Break.pdf

我想问个问题，请问xuehe和zhaomn200145两位大侠会吗？

在Perron（1989）的文章里P1375，它的模拟方法介绍了这么一句"we generate a sample of size 1000 of i.i.d N(0,1) random deviates {et},we then construct sample moments of the data which converge weakly to the various functionals of the wiener proces involved in representation of the asymptotic distributions" .

这句话是不是先产生独立标准正态分布的序列{et},然后再生成各种检验统计量表达式中的维纳过程，计算统计量的值，而不象标准单位根检验的模拟那样直接提取t统计量或者计算统计量。

我用Eviews编的小程序

Perron（1989）P1368-1370图4和图5的模拟程序

回答xuelida：就我的理解，Perron原文中的意思好像是你翻的这样，但是不知道他是怎么编程来实现的，本人对编程这一块还是入门，实在惭愧。以后还要多多向你请教啊！！

i.i.d独立同分布,asymptotic渐近分布,还是极限分布?我也看的稀里糊涂,须看上下文,老外就是厉害,向你学习.

Undefined symbols:
lrvar D:\x\LMUNIT\AL.txt(171)

Undefined symbols:
lrvar D:\x\LMUNIT\LS.txt(241)

在gauss上为不可识别符号?

这两个一定能用，我刚才又测试了一次，能用。你的情况我也不知道错在哪儿。

对于这个问题我又看了他的1990年的文章（Testing for a unit root in a time series with a changing mean），他的模拟程序是这样的，首先对于有限样本还是象以前那样模拟，这个我编了一个小程序eviews（1989），计算模型A的分位数，模型B和C可以自己修改一下就行了。而对于无限样本的分位数就需要利用单位根统计量的形式，根据维纳过程模拟了，这个时候就需要利用独立标准正态分布的序列{et}构造维纳过程，然后计算统计量。

具体可以看Perron（1990）p157-158表4和表5下面的注释

[此贴子已经被作者于2007-8-11 11:17:08编辑过]

这个是上贴的程序Perron（1989），

Perron（1990）Testing for a unit root in a time series with a changing mean

这篇文章的程序，也可以参考上面的。

[此贴子已经被作者于2007-8-11 11:16:13编辑过]

unit检验等程序的.

[此贴子已经被作者于2007-8-18 20:27:01编辑过]

xuehe同志似乎对Gauss不是很熟悉。呵呵

这里程序都能用，我都测试过了