<p>I Think cross spectral methods (the phase spectrum in particular) is</p>
<p>very powerfull. See the working paper 'Spectral Analysis for</p>
<p>Economic Time Series' by Alessandra Iacobucci avalaible</p>
<p>in the web.</p>
<p> </p>
<p>Hope this helps</p>
<p> </p>
<p>Washington Santos Silva<br />Diretoria de Graduação e Pós-Graduação<br />CEFET-Bambuí/MG<br />Tel.: (37)3431:4922<br />e-mail: wss@cefetbambui.edu.br</p>
<p><br /><br />Em 22/01/2009 02:38, <strong><span title="Matthieu Stigler<matthieu.stigler@gmail.com>">Matthieu Stigler </span></strong> escreveu:</p>
<blockquote style="border-left: 2px solid #6868cc; margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br />Michael a écrit :<br />> Hi all,<br />><br />> Is there a way to study the lead and lag relation of two time series?<br />><br />> Let's say I have two time series, At and Bt. Is there a systematic way<br />> of concluding whether it's A leading B or B leading A and by how much?<br />><br />> Thanks!<br />><br />> _______________________________________________<br />> R-SIG-Finance@stat.math.ethz.ch mailing list<br />> https://stat.ethz.ch/mailman/listinfo/r-sig-finance<br />> -- Subscriber-posting only.<br />> -- If you want to post, subscribe first.<br />> <br />You can use cross-correlation:<br />a<-rnorm(100)<br /> > b<-runif(100)<br /> > ccf<-ccf(a,b)<br />plot(ccf)<br />ccf$acf<br /><br />Or use some VAR from package vars.<br /><br />Bests Mat<br /><br />_______________________________________________<br /
>R-SIG-Finance@stat.math.ethz.ch mailing list<br />https://stat.ethz.ch/mailman/listinfo/r-sig-finance<br />-- Subscriber-posting only.<br />-- If you want to post, subscribe first.<br /><br /></blockquote>