[R] Fwd: Use of R for VECM

[R. Michael Weylandt <michael.weylandt@gmail.com>]

I suspect that part of the hesitancy in replying to your query is that your academic e-mail and subsequent google-ability suggest you are a student in the economics department and that this might be homework. 

In the meanwhile, you might want to look into Pfaff's book on cointegration and time series analysis in R. 

Michael

On Nov 11, 2011, at 10:06 AM, vramaiah at neo.tamu.edu wrote:

> 
> ----- Forwarded Message -----
> From: vramaiah at neo.tamu.edu
> To: "bernhard pfaff" <bernhard.pfaff at pfaffikus.de>
> Sent: Friday, November 11, 2011 9:03:11 AM GMT -06:00 US/Canada Central
> Subject: Use of R for VECM
> 
> Hello Fellow R'ers
> I am a new user of R and I am applying it for solving Bi-Variate (Consumption and Output) VECM with Co-Integration (I(1)) with three lags on Consumption and Output expressed aa lags of differences.  R Code and one segment fo the output (other parts of the output are repeatitive) are as follows.
>> us=read.table("usdata.1995-2009.txt",header=T)
>> sjd<-us[,c("Y","C")]
>> sjd.vecm1 <- ca.jo(sjd, ecdet='const', type="eigen", K=3, spec="longrun",
> + season=4)
>> sjd.vecm2 <- ca.jo(sjd, ecdet='const', type="eigen", K=3, spec="transitory",
> + season=4)
>> sjd.vecm.ols1 <- cajools(sjd.vecm1)
>> sjd.vecm.ols2 <- cajools(sjd.vecm2)
>> summary(sjd.vecm.ols1)
> Response Y.d :
> 
> Call:
> lm(formula = substitute(Y.d), data = data.mat)
> 
> Residuals:
>       Min         1Q     Median         3Q        Max 
> -0.0049787 -0.0012948  0.0000703  0.0009653  0.0063192 
> 
> Coefficients:
>           Estimate Std. Error t value Pr(>|t|)    
> sd1      -0.0002012  0.0007653  -0.263 0.793820    
> sd2       0.0013339  0.0007616   1.752 0.086378 .  
> sd3       0.0007372  0.0007947   0.928 0.358348    
> Y.dl1    -0.2215246  0.1600532  -1.384 0.172875    
> C.dl1     0.9220846  0.1646314   5.601 1.08e-06 ***
> Y.dl2    -0.1600219  0.1273838  -1.256 0.215245    
> C.dl2     0.4712112  0.2124175   2.218 0.031401 *  
> Y.l3     -0.3227708  0.0881079  -3.663 0.000631 ***
> C.l3      0.2376579  0.0688854   3.450 0.001194 ** 
> constant  0.4707624  0.1182284   3.982 0.000236 ***
> ---
> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 
> 
> Residual standard error: 0.002013 on 47 degrees of freedom
> Multiple R-squared: 0.7053,    Adjusted R-squared: 0.6426 
> F-statistic: 11.25 on 10 and 47 DF,  p-value: 1.707e-09 
> 
> 
> 
> I am trying to decipher the output and have the following questions.  
> 1.  If you have published a book with explanations of the output, I would like to acquire one.  Please advise me where I can get one.  It might me and you lot of time.  In the meantime....
> 2.  Are Y.dl1, C.dl1, Y.dl2, C.dl2, Y.l3 and C.l3 are the coefficients I am looking for?
> 3.  My equition requires Y.l1 and C.l1.  Is there a way I could change from Y.l3 and C.l3 
> 4.  I need to use these coefficients for forecasting.  What command I use to extract the coefficients from this program?
> 5.  What are sd1, sd2, sd3?  I did not input any "seasonal dummies" and yet the output seems to have included them.  Does it affect the value of the coefficients?  Is there a better way to reframe the command to avoid them?
> I hope it is not too much to ask for your help.
> I thank you in advance.
> Ram Ramaiah
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.