wilmott.com

Hello,How do I interpret Johansen's test results. I have an application calculating using the Johansen's Test but I do not understand the results, so I would be very happy if someone could explain it to me.Thanks!

These are the results. How do I interpret this fascinating code;> summary(ca.jo(cbind(zPEO, zXLE),type = "eigen", K = 2)) 1. 1. Johansen-Procedure # 1.Test type: maximal eigenvalue statistic (lambda max) , with linear trendEigenvalues (lambda):[1] 0.0081571415 0.0006444724Values of teststatistic and critical values of test: test 10pct 5pct 1pctr <= 1 | 1.62 6.50 8.18 11.65r = 0 | 20.57 12.91 14.90 19.19Eigenvectors, normalised to first column:(These are the cointegration relations) zPEO.l2 zXLE.l2zPEO.l2 1.0000000 1.000000zXLE.l2 -0.3484042 -1.749159Weights W:(This is the loading matrix) zPEO.l2 zXLE.l2zPEO.d -0.009555347 0.0003349528zXLE.d 0.044504843 0.0001818289> summary(ca.jo(cbind(zPEO, zXLE),type = "trace", K = 2)) 1. 1. Johansen-Procedure # 1.Test type: trace statistic , with linear trendEigenvalues (lambda):[1] 0.0081571415 0.0006444724Values of teststatistic and critical values of test: test 10pct 5pct 1pctr <= 1 | 1.62 6.50 8.18 11.65r = 0 | 22.19 15.66 17.95 23.52Eigenvectors, normalised to first column:(These are the cointegration relations) zPEO.l2 zXLE.l2zPEO.l2 1.0000000 1.000000zXLE.l2 -0.3484042 -1.749159Weights W:(This is the loading matrix) zPEO.l2 zXLE.l2zPEO.d -0.009555347 0.0003349528zXLE.d 0.044504843 0.0001818289

I also have a Excel app doing this calculation if someone would like to see it in action before you are able to interpret the results.

Nobody understand Johansen's test? Or nobody want to help out?

I've found a very nice FAQ which should help you.http://www.quantcode.com/modules/smartf ... ?faqid=103

For this particular R code I think the best source is the book Analysis of Integrated and Cointegrated Time Series with R by Pfaff.If you are in r just type ?ca.jo or look in google for the manual of the package urca.If you want a nice introduction to cointegration look at textbooks in time series econometrics such as Enders, Tsay, or Zivot.Also you might want to take a look to these papers by Hendry and Juselius:http://www.econ.ku.dk/okokj/papers/dfhk ... hkjfnl.pdf