<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"> <!--Converted with LaTeX2HTML 98.1p1 release (March 2nd, 1998) originally by Nikos Drakos (nikos@cbl.leeds.ac.uk), CBLU, University of Leeds * revised and updated by: Marcus Hennecke, Ross Moore, Herb Swan * with significant contributions from: Jens Lippmann, Marek Rouchal, Martin Wilck and others --> <HTML> <HEAD> <TITLE>The power spectrum and covariance statistics</TITLE> <META NAME="description" CONTENT="The power spectrum and covariance statistics"> <META NAME="keywords" CONTENT="vol2"> <META NAME="resource-type" CONTENT="document"> <META NAME="distribution" CONTENT="global"> <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1"> <LINK REL="STYLESHEET" HREF="vol2.css"> <LINK REL="next" HREF="node234.html"> <LINK REL="previous" HREF="node232.html"> <LINK REL="up" HREF="node231.html"> <LINK REL="next" HREF="node234.html"> </HEAD> <BODY > <!--Navigation Panel--> <A NAME="tex2html4315" HREF="node234.html"> <IMG WIDTH="37" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="next" SRC="icons.gif/next_motif.gif"></A> <A NAME="tex2html4312" HREF="node231.html"> <IMG WIDTH="26" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="up" SRC="icons.gif/up_motif.gif"></A> <A NAME="tex2html4306" HREF="node232.html"> <IMG WIDTH="63" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="previous" SRC="icons.gif/previous_motif.gif"></A> <A NAME="tex2html4314" HREF="node1.html"> <IMG WIDTH="65" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="contents" SRC="icons.gif/contents_motif.gif"></A> <BR> <B> Next:</B> <A NAME="tex2html4316" HREF="node234.html">Sampling patterns</A> <B> Up:</B> <A NAME="tex2html4313" HREF="node231.html">Fourier analysis: The sine</A> <B> Previous:</B> <A NAME="tex2html4307" HREF="node232.html">Fourier transforms</A> <BR> <BR> <!--End of Navigation Panel--> <H2><A NAME="SECTION001732000000000000000"> </A><A NAME="s:psac"> </A> <BR> The power spectrum and covariance statistics </H2> <P> Let us define power spectrum, covariance and autocovariance statistics <I>P</I>, <I>Cov</I> and <I>ACF</I>: <BR> <DIV ALIGN="CENTER"> <!-- MATH: \begin{eqnarray} P[x](\nu) & = & |{\cal F}x|^2 \\ Cov[x,y](l) & = & x(t)*y(-t)\\ ACF[x](l) & = & Cov[x,x](l) \end{eqnarray} --> <TABLE ALIGN="CENTER" CELLPADDING="0" WIDTH="100%"> <TR VALIGN="MIDDLE"><TD NOWRAP ALIGN="RIGHT"><IMG WIDTH="32" HEIGHT="42" ALIGN="MIDDLE" BORDER="0" SRC="img509.gif" ALT="$\displaystyle P[x](\nu)$"></TD> <TD ALIGN="CENTER" NOWRAP>=</TD> <TD ALIGN="LEFT" NOWRAP><IMG WIDTH="12" HEIGHT="50" ALIGN="MIDDLE" BORDER="0" SRC="img510.gif" ALT="$\displaystyle \vert{\cal F}x\vert^2$"></TD> <TD WIDTH=10 ALIGN="RIGHT"> (12.7)</TD></TR> <TR VALIGN="MIDDLE"><TD NOWRAP ALIGN="RIGHT"><I>Cov</I>[<I>x</I>,<I>y</I>](<I>l</I>)</TD> <TD ALIGN="CENTER" NOWRAP>=</TD> <TD ALIGN="LEFT" NOWRAP><I>x</I>(<I>t</I>)*<I>y</I>(-<I>t</I>)</TD> <TD WIDTH=10 ALIGN="RIGHT"> (12.8)</TD></TR> <TR VALIGN="MIDDLE"><TD NOWRAP ALIGN="RIGHT"><I>ACF</I>[<I>x</I>](<I>l</I>)</TD> <TD ALIGN="CENTER" NOWRAP>=</TD> <TD ALIGN="LEFT" NOWRAP><I>Cov</I>[<I>x</I>,<I>x</I>](<I>l</I>)</TD> <TD WIDTH=10 ALIGN="RIGHT"> (12.9)</TD></TR> </TABLE></DIV> <BR CLEAR="ALL"><P></P> The power spectrum is special among the periodograms in that it is the square of a linear operator and reveals the important correspondence between frequency and time domain analyses: <BR> <DIV ALIGN="CENTER"> <!-- MATH: \begin{eqnarray} P[x](\nu) & = & {\cal F}[ACF[x](l)](\nu), \end{eqnarray} --> <TABLE ALIGN="CENTER" CELLPADDING="0" WIDTH="100%"> <TR VALIGN="MIDDLE"><TD NOWRAP ALIGN="RIGHT"><IMG WIDTH="32" HEIGHT="42" ALIGN="MIDDLE" BORDER="0" SRC="img511.gif" ALT="$\displaystyle P[x](\nu)$"></TD> <TD ALIGN="CENTER" NOWRAP>=</TD> <TD ALIGN="LEFT" NOWRAP><IMG WIDTH="124" HEIGHT="44" ALIGN="MIDDLE" BORDER="0" SRC="img512.gif" ALT="$\displaystyle {\cal F}[ACF[x](l)](\nu),$"></TD> <TD WIDTH=10 ALIGN="RIGHT"> (12.10)</TD></TR> </TABLE></DIV> <BR CLEAR="ALL"><P></P> by virtue of Eq. (<A HREF="node232.html#e:fcon">12.5</A>). <P> Let us consider which linear operators or matrices convert series of independent random variables into series of independent variables. For the discrete, evenly sampled observations the ACF is computed as the scalar product of vectors obtained by circularly permutating the data of the series. For a series of independent random variables, e.g. white noise, the vectors are orthogonal. It is known from linear algebra that only orthogonal matrices preserve orthogonality. So, only in the special case of evenly spaced discrete observations and frequencies (Sect. <A HREF="node232.html#s:iftr">12.3.1</A>) are <!-- MATH: ${\cal F}[x]$ --> <IMG WIDTH="72" HEIGHT="74" ALIGN="MIDDLE" BORDER="0" SRC="img513.gif" ALT="${\cal F}[x]$"> (and <I>P</I>[<I>x</I>]) independent for each frequency. In the next subsection we discuss the case of dependent and correlated values of <I>P</I>[<I>x</I>]. <P> <HR> <!--Navigation Panel--> <A NAME="tex2html4315" HREF="node234.html"> <IMG WIDTH="37" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="next" SRC="icons.gif/next_motif.gif"></A> <A NAME="tex2html4312" HREF="node231.html"> <IMG WIDTH="26" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="up" SRC="icons.gif/up_motif.gif"></A> <A NAME="tex2html4306" HREF="node232.html"> <IMG WIDTH="63" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="previous" SRC="icons.gif/previous_motif.gif"></A> <A NAME="tex2html4314" HREF="node1.html"> <IMG WIDTH="65" HEIGHT="24" ALIGN="BOTTOM" BORDER="0" ALT="contents" SRC="icons.gif/contents_motif.gif"></A> <BR> <B> Next:</B> <A NAME="tex2html4316" HREF="node234.html">Sampling patterns</A> <B> Up:</B> <A NAME="tex2html4313" HREF="node231.html">Fourier analysis: The sine</A> <B> Previous:</B> <A NAME="tex2html4307" HREF="node232.html">Fourier transforms</A> <!--End of Navigation Panel--> <ADDRESS> <I>Petra Nass</I> <BR><I>1999-06-15</I> </ADDRESS> </BODY> </HTML>