s-news
[Top] [All Lists]

Cointegration procedure

To: <s-news@lists.biostat.wustl.edu>
Subject: Cointegration procedure
From: "Jan Verbesselt" <Jan.Verbesselt@agr.kuleuven.ac.be>
Date: Mon, 24 May 2004 16:00:30 +0200
Cc: "'prad s u'" <prad@mail.ahc.umn.edu>
Importance: Normal

Hi S helpers,

 

To check how two time series (I(1), non-stationary, show seasonal variation), are related/correlated with each-other I started looking beside correlation coefficients at 'cointegration' possibilities.

 

How can the strength of coïntegration be assessed?  I’m using the book Modelling Fin. Time series with S-plus book but get stuck each example. Is coïntegration appropriate in this case and is there a good method to assess the strength of it?

The B2 coefficient is different from 1 so the DOLS estimator of B2 has to be used. How can this be done is this case (12.4 p435)?

 

a)    deseasonalise the data by diff? is than the real coïntegration assessed?

b)    DOLS(…)? Regression equation estimation? How can the following be solved (correct) towards cointegration vector estimation?

 

attach(Timeserie)

serie1    <- rts(Timeserie[,1],start= c(1998,10), frequency=36)

serie2    <- rts(Timeserie[,2],start= c(1998,10), frequency=36)

detach(Timeserie)

#(regular spaced time series, non calendar based

ts.ser1    <-ts.update(serie1)

ts.ser2    <-ts.update(serie2)

 

> ts.ser1d                <- diff(ts.ser1) # signalSeries

> colIds(ts.ser1d )      <- c( "D.serie1")

Warning messages:

  Cannot set a column ID for a vector in: colIds(.A0, .A1)

> ts.ser1.dlag            <- tslag(ts.ser1d ,-3:3,trim=T)

> merge.DOLS              <- seriesMerge(ts.ser2,ts.ser1,ts.ser1.dlag)

Problem in object@positions: Class "matrix" has no "positions"

slot

Use traceback() to see the call stack

>

 

=> S-plus 6.1, module financial metrics 1.0, Win XP station

 

Thanks a lot in advance,

Ph.D. student,

Jan

 

______________________________________________________________________

Jan Verbesselt

Research Associate

Lab of Geomatics and Forest Engineering K.U. Leuven

Vital Decosterstraat 102. B-3000 Leuven Belgium

Tel:+32-16-329750   Fax: +32-16-329760

http://gloveg.kuleuven.ac.be/

_______________________________________________________________________

 

 

Results:

The estimated regression function between the two time series! (OLS)

 

Call:

OLS(formula = ser2 ~ ser1, data = "">

 

Residuals:

       Min        1Q    Median        3Q       Max

 -212.8883  -63.4366    9.4245   68.1576  315.3256

 

Coefficients:

                 Value Std. Error    t value   Pr(>|t|)

(Intercept)  -483.1393     8.8140   -54.8148     0.0000

       ser1  1519.7852    56.7010    26.8035     0.0000

 

Regression Diagnostics:

                         

         R-Squared 0.8023

Adjusted R-Squared 0.8012

Durbin-Watson Stat 0.8990

 

Residual Diagnostics:

                Stat  P-Value

Jarque-Bera   1.4695   0.4796

  Ljung-Box 104.6771   0.0000

 

Residual standard error: 91.12 on 177 degrees of freedom

F-statistic: 718.4 on 1 and 177 degrees of freedom, the p-value

 is 0

(strong autocorrelation indicated by the DW test and ljung-box)

 

> us  <- serie1 - serie2

> unitroot(us,trend="c",method="adf",lags=11)

 

 

Test for Unit Root: Augmented DF Test

 

Null Hypothesis: there is a unit root

   Type of Test: t-test

 Test Statistic: -4.831

        P-value: 7.871e-5

 

Coefficients:

     lag1     lag2     lag3     lag4     lag5     lag6

  -0.1917   0.2523   0.0920   0.1753   0.0883   0.0728

 

     lag7     lag8     lag9    lag10    lag11 constant

   0.0147   0.1847   0.0730   0.0991   0.1218  62.4556

 

Degrees of freedom: 168 total; 156 residual

Residual standard error: 71.94 _

<Prev in Thread] Current Thread [Next in Thread>
  • Cointegration procedure, Jan Verbesselt <=