1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
|
<html lang="en">
<head>
<title>Famous Matrices - GNU Octave</title>
<meta http-equiv="Content-Type" content="text/html">
<meta name="description" content="GNU Octave">
<meta name="generator" content="makeinfo 4.13">
<link title="Top" rel="start" href="index.html#Top">
<link rel="up" href="Matrix-Manipulation.html#Matrix-Manipulation" title="Matrix Manipulation">
<link rel="prev" href="Special-Utility-Matrices.html#Special-Utility-Matrices" title="Special Utility Matrices">
<link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
<meta http-equiv="Content-Style-Type" content="text/css">
<style type="text/css"><!--
pre.display { font-family:inherit }
pre.format { font-family:inherit }
pre.smalldisplay { font-family:inherit; font-size:smaller }
pre.smallformat { font-family:inherit; font-size:smaller }
pre.smallexample { font-size:smaller }
pre.smalllisp { font-size:smaller }
span.sc { font-variant:small-caps }
span.roman { font-family:serif; font-weight:normal; }
span.sansserif { font-family:sans-serif; font-weight:normal; }
--></style>
</head>
<body>
<div class="node">
<a name="Famous-Matrices"></a>
<p>
Previous: <a rel="previous" accesskey="p" href="Special-Utility-Matrices.html#Special-Utility-Matrices">Special Utility Matrices</a>,
Up: <a rel="up" accesskey="u" href="Matrix-Manipulation.html#Matrix-Manipulation">Matrix Manipulation</a>
<hr>
</div>
<h3 class="section">16.4 Famous Matrices</h3>
<p>The following functions return famous matrix forms.
<!-- hadamard scripts/special-matrix/hadamard.m -->
<p><a name="doc_002dhadamard"></a>
<div class="defun">
— Function File: <b>hadamard</b> (<var>n</var>)<var><a name="index-hadamard-1761"></a></var><br>
<blockquote><p>Construct a Hadamard matrix (Hn) of size <var>n</var>-by-<var>n</var>. The
size <var>n</var> must be of the form 2^k * p in which
p is one of 1, 12, 20 or 28. The returned matrix is normalized,
meaning <code>Hn(:,1) == 1</code><!-- /@w --> and <code>Hn(1,:) == 1</code><!-- /@w -->.
<p>Some of the properties of Hadamard matrices are:
<ul>
<li><code>kron (Hm, Hn)</code> is a Hadamard matrix of size <var>m</var>-by-<var>n</var>.
<li><code>Hn * Hn' = </code><var>n</var><code> * eye (</code><var>n</var><code>)</code>.
<li>The rows of Hn are orthogonal.
<li><code>det (</code><var>A</var><code>) <= abs (det (Hn))</code> for all <var>A</var> with
<code>abs (</code><var>A</var><code>(i, j)) <= 1</code><!-- /@w -->.
<li>Multiplying any row or column by -1 and the matrix will remain a Hadamard
matrix.
</ul>
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dcompan.html#doc_002dcompan">compan</a>, <a href="doc_002dhankel.html#doc_002dhankel">hankel</a>, <a href="doc_002dtoeplitz.html#doc_002dtoeplitz">toeplitz</a>.
</p></blockquote></div>
<!-- hankel scripts/special-matrix/hankel.m -->
<p><a name="doc_002dhankel"></a>
<div class="defun">
— Function File: <b>hankel</b> (<var>c</var>)<var><a name="index-hankel-1762"></a></var><br>
— Function File: <b>hankel</b> (<var>c, r</var>)<var><a name="index-hankel-1763"></a></var><br>
<blockquote><p>Return the Hankel matrix constructed from the first column <var>c</var>, and
(optionally) the last row <var>r</var>. If the last element of <var>c</var> is
not the same as the first element of <var>r</var>, the last element of
<var>c</var> is used. If the second argument is omitted, it is assumed to
be a vector of zeros with the same size as <var>c</var>.
<p>A Hankel matrix formed from an m-vector <var>c</var>, and an n-vector
<var>r</var>, has the elements
<pre class="example"> H(i,j) = c(i+j-1), i+j-1 <= m;
H(i,j) = r(i+j-m), otherwise
</pre>
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dhadamard.html#doc_002dhadamard">hadamard</a>, <a href="doc_002dtoeplitz.html#doc_002dtoeplitz">toeplitz</a>.
</p></blockquote></div>
<!-- hilb scripts/special-matrix/hilb.m -->
<p><a name="doc_002dhilb"></a>
<div class="defun">
— Function File: <b>hilb</b> (<var>n</var>)<var><a name="index-hilb-1764"></a></var><br>
<blockquote><p>Return the Hilbert matrix of order <var>n</var>. The i,j element
of a Hilbert matrix is defined as
<pre class="example"> H(i, j) = 1 / (i + j - 1)
</pre>
<p>Hilbert matrices are close to being singular which make them difficult to
invert with numerical routines.
Comparing the condition number of a random matrix 5x5 matrix with that of
a Hilbert matrix of order 5 reveals just how difficult the problem is.
<pre class="example"> cond (rand (5))
⇒ 14.392
cond (hilb (5))
⇒ 4.7661e+05
</pre>
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dinvhilb.html#doc_002dinvhilb">invhilb</a>.
</p></blockquote></div>
<!-- invhilb scripts/special-matrix/invhilb.m -->
<p><a name="doc_002dinvhilb"></a>
<div class="defun">
— Function File: <b>invhilb</b> (<var>n</var>)<var><a name="index-invhilb-1765"></a></var><br>
<blockquote><p>Return the inverse of the Hilbert matrix of order <var>n</var>. This can be
computed exactly using
<pre class="example">
(i+j) /n+i-1\ /n+j-1\ /i+j-2\ 2
A(i,j) = -1 (i+j-1)( )( ) ( )
\ n-j / \ n-i / \ i-2 /
= p(i) p(j) / (i+j-1)
</pre>
<p class="noindent">where
<pre class="example"> k /k+n-1\ /n\
p(k) = -1 ( ) ( )
\ k-1 / \k/
</pre>
<p>The validity of this formula can easily be checked by expanding
the binomial coefficients in both formulas as factorials. It can
be derived more directly via the theory of Cauchy matrices.
See J. W. Demmel, <cite>Applied Numerical Linear Algebra</cite>, p. 92.
<p>Compare this with the numerical calculation of <code>inverse (hilb (n))</code>,
which suffers from the ill-conditioning of the Hilbert matrix, and the
finite precision of your computer's floating point arithmetic.
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dhilb.html#doc_002dhilb">hilb</a>.
</p></blockquote></div>
<!-- magic scripts/special-matrix/magic.m -->
<p><a name="doc_002dmagic"></a>
<div class="defun">
— Function File: <b>magic</b> (<var>n</var>)<var><a name="index-magic-1766"></a></var><br>
<blockquote>
<p>Create an <var>n</var>-by-<var>n</var> magic square. A magic square is an arrangement
of the integers <code>1:n^2</code> such that the row sums, column sums, and
diagonal sums are all equal to the same value.
<p>Note: <var>n</var> must be greater than 2 for the magic square to exist.
</p></blockquote></div>
<!-- pascal scripts/special-matrix/pascal.m -->
<p><a name="doc_002dpascal"></a>
<div class="defun">
— Function File: <b>pascal</b> (<var>n</var>)<var><a name="index-pascal-1767"></a></var><br>
— Function File: <b>pascal</b> (<var>n, t</var>)<var><a name="index-pascal-1768"></a></var><br>
<blockquote><p>Return the Pascal matrix of order <var>n</var> if <var>t</var><code> = 0</code>. <var>t</var>
defaults to 0. Return the pseudo-lower triangular Cholesky factor of
the Pascal matrix if <var>t</var><code> = 1</code> (The sign of some columns may be
negative). This matrix is its own inverse, that is <code>pascal (</code><var>n</var><code>,
1) ^ 2 == eye (</code><var>n</var><code>)</code>. If <var>t</var><code> = -1</code>, return the true
Cholesky factor with strictly positive values on the diagonal. If
<var>t</var><code> = 2</code>, return a transposed and permuted version of <code>pascal
(</code><var>n</var><code>, 1)</code>, which is the cube root of the identity matrix. That is,
<code>pascal (</code><var>n</var><code>, 2) ^ 3 == eye (</code><var>n</var><code>)</code>.
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dchol.html#doc_002dchol">chol</a>.
</p></blockquote></div>
<!-- rosser scripts/special-matrix/rosser.m -->
<p><a name="doc_002drosser"></a>
<div class="defun">
— Function File: <b>rosser</b> ()<var><a name="index-rosser-1769"></a></var><br>
<blockquote><p>Return the Rosser matrix. This is a difficult test case used to evaluate
eigenvalue algorithms.
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dwilkinson.html#doc_002dwilkinson">wilkinson</a>, <a href="doc_002deig.html#doc_002deig">eig</a>.
</p></blockquote></div>
<!-- toeplitz scripts/special-matrix/toeplitz.m -->
<p><a name="doc_002dtoeplitz"></a>
<div class="defun">
— Function File: <b>toeplitz</b> (<var>c</var>)<var><a name="index-toeplitz-1770"></a></var><br>
— Function File: <b>toeplitz</b> (<var>c, r</var>)<var><a name="index-toeplitz-1771"></a></var><br>
<blockquote><p>Return the Toeplitz matrix constructed from the first column <var>c</var>,
and (optionally) the first row <var>r</var>. If the first element of <var>r</var>
is not the same as the first element of <var>c</var>, the first element of
<var>c</var> is used. If the second argument is omitted, the first row is
taken to be the same as the first column.
<p>A square Toeplitz matrix has the form:
<pre class="example"> c(0) r(1) r(2) ... r(n)
c(1) c(0) r(1) ... r(n-1)
c(2) c(1) c(0) ... r(n-2)
. . . . .
. . . . .
. . . . .
c(n) c(n-1) c(n-2) ... c(0)
</pre>
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dhankel.html#doc_002dhankel">hankel</a>.
</p></blockquote></div>
<!-- vander scripts/special-matrix/vander.m -->
<p><a name="doc_002dvander"></a>
<div class="defun">
— Function File: <b>vander</b> (<var>c</var>)<var><a name="index-vander-1772"></a></var><br>
— Function File: <b>vander</b> (<var>c, n</var>)<var><a name="index-vander-1773"></a></var><br>
<blockquote><p>Return the Vandermonde matrix whose next to last column is <var>c</var>.
If <var>n</var> is specified, it determines the number of columns;
otherwise, <var>n</var> is taken to be equal to the length of <var>c</var>.
<p>A Vandermonde matrix has the form:
<pre class="example"> c(1)^(n-1) ... c(1)^2 c(1) 1
c(2)^(n-1) ... c(2)^2 c(2) 1
. . . . .
. . . . .
. . . . .
c(n)^(n-1) ... c(n)^2 c(n) 1
</pre>
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002dpolyfit.html#doc_002dpolyfit">polyfit</a>.
</p></blockquote></div>
<!-- wilkinson scripts/special-matrix/wilkinson.m -->
<p><a name="doc_002dwilkinson"></a>
<div class="defun">
— Function File: <b>wilkinson</b> (<var>n</var>)<var><a name="index-wilkinson-1774"></a></var><br>
<blockquote><p>Return the Wilkinson matrix of order <var>n</var>. Wilkinson matrices are
symmetric and tridiagonal with pairs of nearly, but not exactly, equal
eigenvalues. They are useful in testing the behavior and performance
of eigenvalue solvers.
<!-- Texinfo @sp should work but in practice produces ugly results for HTML. -->
<!-- A simple blank line produces the correct behavior. -->
<!-- @sp 1 -->
<p class="noindent"><strong>See also:</strong> <a href="doc_002drosser.html#doc_002drosser">rosser</a>, <a href="doc_002deig.html#doc_002deig">eig</a>.
</p></blockquote></div>
<!-- DO NOT EDIT! Generated automatically by munge-texi. -->
<!-- Copyright (C) 1996-2012 John W. Eaton -->
<!-- This file is part of Octave. -->
<!-- Octave is free software; you can redistribute it and/or modify it -->
<!-- under the terms of the GNU General Public License as published by the -->
<!-- Free Software Foundation; either version 3 of the License, or (at -->
<!-- your option) any later version. -->
<!-- Octave is distributed in the hope that it will be useful, but WITHOUT -->
<!-- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or -->
<!-- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License -->
<!-- for more details. -->
<!-- You should have received a copy of the GNU General Public License -->
<!-- along with Octave; see the file COPYING. If not, see -->
<!-- <http://www.gnu.org/licenses/>. -->
</body></html>
|
