Re: [R] Problems with Cochrane-Orcutt procedures

From: <tolga.i.uzuner_at_jpmorgan.com>
Date: Tue, 17 Jun 2008 18:17:03 +0100


Sure, I can imagine GLS is a much better way to deal with this.

I guess I was looking at this because I did try GLS but got exactly the same results as LM and I just wanted to be sure.

I did "debug" the code in
https://stat.ethz.ch/pipermail/r-help/2002-January/017774.html and the offending line is:

...
mod <- lm(y ~ X - 1)
...

in the Orcutt procedure. Essentially, X is twice the length of y, as below:

> cochrane.orcutt(regrCMSlm)
debugging in: cochrane.orcutt(regrCMSlm) debug: {

    UseMethod("cochrane.orcutt")
}
Browse[1]>
debug: UseMethod("cochrane.orcutt")
Browse[1]>
debugging in: cochrane.orcutt.lm(regrCMSlm) debug: {

    X <- model.matrix(mod)
    y <- model.response(model.frame(mod))
    e <- residuals(mod)
    n <- length(e)

    names <- colnames(X)
    rho <- sum(e[1:(n - 1)] * e[2:n])/sum(e^2)     y <- y[2:n] - rho * y[1:(n - 1)]
    X <- X[2:n, ] - rho * X[1:(n - 1), ]     mod <- lm(y ~ X - 1)
    result <- list()
    result$coefficients <- coef(mod)
    names(result$coefficients) <- names
    summary <- summary(mod, corr = F)
    result$cov <- (summary$sigma^2) * summary$cov.unscaled     dimnames(result$cov) <- list(names, names)     result$sigma <- summary$sigma
    result$rho <- rho
    class(result) <- "cochrane.orcutt"
    result
}
Browse[1]>
debug: X <- model.matrix(mod)
Browse[1]>
debug: y <- model.response(model.frame(mod)) Browse[1]> length(X)
[1] 378 #<<<<<<<<<<<<<<<<------------------------------
Browse[1]>
debug: e <- residuals(mod)
Browse[1]> length(y)
[1] 189 #<<<<<<<<<<<<<<<<------------------------------
Browse[1]>
debug: n <- length(e)
Browse[1]>
debug: names <- colnames(X)
Browse[1]>
debug: rho <- sum(e[1:(n - 1)] * e[2:n])/sum(e^2) Browse[1]>
debug: y <- y[2:n] - rho * y[1:(n - 1)]
Browse[1]>
debug: X <- X[2:n, ] - rho * X[1:(n - 1), ] Browse[1]>
debug: mod <- lm(y ~ X - 1)
Browse[1]>
Error in model.frame.default(formula = y ~ X - 1, drop.unused.levels = TRUE) :
  variable lengths differ (found for 'X')

Tolga

"John Fox" <jfox_at_mcmaster.ca>
17/06/2008 17:51

To
<tolga.i.uzuner_at_jpmorgan.com>
cc
<r-help_at_r-project.org>
Subject
RE: [R] Problems with Cochrane-Orcutt procedures

Dear Tolga,

I'm afraid that I don't see an error. (I expect in any event that the Cochrane-Orcott and Prais estimators are now only of historical interest.)

Regards,
 John



John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
web: socserv.mcmaster.ca/jfox

> -----Original Message-----
> From: r-help-bounces_at_r-project.org [mailto:r-help-bounces_at_r-project.org]
On
> Behalf Of tolga.i.uzuner_at_jpmorgan.com
> Sent: June-17-08 11:13 AM
> To: r-help@r-project.org
> Subject: [R] Problems with Cochrane-Orcutt procedures
>
> Hi John,
>
> Hi Folks/Prof. Fox,
>
> I found some code John Fox had written sometime back on the
Cochrane-Orcutt
> and Prais procedures here:
> https://stat.ethz.ch/pipermail/r-help/2002-January/017774.html
>
> I thought I would try it out and get the following errors below. Was
> wondering if anyone had any immediate opinions why this might be ?
>
> The linear model is the object regrCMSlm .
>
> Thanks,
> Tolga
>
> > regrCMSlm
>
> Call:
> lm(formula = regrCMS[, 1] ~ regrCMS[, 2])
>
> Coefficients:
> (Intercept) regrCMS[, 2]
> 25.7067 -0.3409
>
> > summary(regrCMSlm)
>
> Call:
> lm(formula = regrCMS[, 1] ~ regrCMS[, 2])
>
> Residuals:
> 09/20/07 11/28/07 02/01/08 04/09/08 06/16/08
> 10.0593 0.3588 -1.1459 0.1340 -9.8520
>
> Coefficients:
> Estimate Std. Error t value Pr(>|t|)
> (Intercept) 25.70673 0.85300 30.14 <2e-16 ***
> regrCMS[, 2] -0.34092 0.02205 -15.46 <2e-16 ***
> ---
> Signif. codes: 0 b

Generally, this communication is for informational purposes only and it is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. In the event you are receiving the offering materials attached below related to your interest in hedge funds or private equity, this communication may be intended as an offer or solicitation for the purchase or sale of such fund(s). All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of JPMorgan Chase & Co., its subsidiaries and affiliates.

This transmission may contain information that is privileged, confidential, legally privileged, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. Although this transmission and any attachments are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by JPMorgan Chase & Co., its subsidiaries and affiliates, as applicable, for any loss or damage arising in any way from its use. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you.
Please refer to
http://www.jpmorgan.com/pages/disclosures for disclosures relating to UK legal entities.

        [[alternative HTML version deleted]]



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. Received on Tue 17 Jun 2008 - 17:37:02 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Tue 17 Jun 2008 - 18:31:14 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive