Have you received a reply to this? I haven't seen one. If not, you
might have received a quicker reply if you had noted that the example
came from ch. 14 of Venables and Ripley (2002) Modern applied Statistics
with S, 4th ed. (Springer, pp. 390-391, to be precise).
You ask about the "plot in the lower right-hand corner", which as I
read it is labeled "fdeaths". I assume you are asking why the
upper-right is labeled "mdeaths and fdeaths" and the lower-left
"fdeaths
and mdeaths", especially since the lags in the lower left are negative.
I agree that it might make more sense, since the lower left lags are
negative, to have the label in the lower left the same as in the upper
right. However, it doesn't disturb me greatly as it is.
spencer graves
DeBarr, Dave wrote:> When I run the "acf()" function using the
"acf(ts.union(mdeaths,
> fdeaths))" example, the "acf()" function calls the
"acf.plot()"
> function to generate this plot...
> http://members.cox.net/ddebarr/images/acf_example.png
>
> The plot in the lower right-hand corner is labeled "fdeaths &
mdeaths",
> but the negative lags appear to belong to "mdeaths & fdeaths"
[which
> correspond to the positive lags of "fdeaths & mdeaths"].
>
> Am I missing something, or should the plot in the lower right-hand
> corner be labeled "mdeaths & fdeaths" (instead of
"fdeaths & mdeaths")?
>
> Note: The unit of measure for the lags is years.
>
>
>>autocorrelation <- function(x, y, lags) {
>
> + n <- length(x)
> + x.bar <- mean(x)
> + y.bar <- mean(y)
> + c <- array(0, length(lags))
> + i <- 1
> + for (t in lags) {
> + s <- seq(max(1, 1 - t), min(n - t, n))
> + c[i] <- sum((x[s + t] - x.bar) * (y[s] - y.bar)) / n
> + i <- i + 1
> + }
> + x.sd <- sqrt(sum((x - x.bar) ^ 2) / n)
> + y.sd <- sqrt(sum((y - y.bar) ^ 2) / n)
> + return(c / (x.sd * y.sd))
> + }
>
>>autocorrelation(mdeaths, fdeaths, -15:15)
>
> [1] 0.015054983 0.365626026 0.615427121 0.708206289 0.621895801
> [6] 0.340005447 -0.024534195 -0.381671430 -0.611793479 -0.677803477
> [11] -0.604031174 -0.349468396 0.019759425 0.405200639 0.744309322
> [16] 0.976241251 0.735668532 0.364241839 -0.010675725 -0.382920620
> [21] -0.622386979 -0.688538519 -0.610583980 -0.383338305 -0.018112073
> [26] 0.391983088 0.656592111 0.721397236 0.639104375 0.361352626
> [31] -0.003385423
>
>>autocorrelation(fdeaths, mdeaths, -15:15)
>
> [1] -0.003385423 0.361352626 0.639104375 0.721397236 0.656592111
> [6] 0.391983088 -0.018112073 -0.383338305 -0.610583980 -0.688538519
> [11] -0.622386979 -0.382920620 -0.010675725 0.364241839 0.735668532
> [16] 0.976241251 0.744309322 0.405200639 0.019759425 -0.349468396
> [21] -0.604031174 -0.677803477 -0.611793479 -0.381671430 -0.024534195
> [26] 0.340005447 0.621895801 0.708206289 0.615427121 0.365626026
> [31] 0.015054983
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
--
Spencer Graves, PhD
Senior Development Engineer
PDF Solutions, Inc.
333 West San Carlos Street Suite 700
San Jose, CA 95110, USA
spencer.graves at pdf.com
www.pdf.com <http://www.pdf.com>
Tel: 408-938-4420
Fax: 408-280-7915