線型代数学/行列概論のソースを表示
←
線型代数学/行列概論
ナビゲーションに移動
検索に移動
あなたには「このページの編集」を行う権限がありません。理由は以下の通りです:
この操作は、次のグループに属する利用者のみが実行できます:
登録利用者
。
このページのソースの閲覧やコピーができます。
{{ナビゲーション|本=[[線型代数学]]|前ページ=[[線型代数学/ベクトル|ベクトル]]|ページ名=行列概論|次ページ=[[線型代数学/線型方程式|線型方程式]]}} ==行列== === 定義 === <math>m</math>と<math>n</math>を自然数とする。<math>mn</math>個の数<math>a_{11}, a_{21}, \cdots, a_{m1}, \cdots, a_{mn}</math>を、括弧で囲んだ中に次のように縦に<math>m</math>個、横に<math>n</math>個、表のように並べて書いたものを、<math>m</math>行<math>n</math>列の'''行列'''(matrix)と言う。(m×n)-行列とも言う。 :<math>A = \begin{pmatrix} a_{11} & a_{12} & a_{13} & \dots & a_{1n}\\ a_{21} & a_{22} & a_{23} & \dots & a_{2n}\\ a_{31} & a_{32} & a_{33} & \dots & a_{3n}\\ \vdots & \vdots & \vdots & \ddots& \vdots\\ a_{m1} & a_{m2} & a_{m3} & \dots & a_{mn}\\ \end{pmatrix}</math> この行列を構成する<math>a_{ij}</math>を行列の'''成分'''(element)と言う。横に並んだ一列を'''行'''(row)、縦に並んだ一列を'''列'''(column)と言う。上から<math>i</math>番目の行を第<math>i</math>行といい、左から<math>j</math>番目の列を第<math>j</math>列と言う。行列内の第<math>i</math>行、第<math>j</math>列に位置する成分を、この行列の<math>(i,j)</math>-成分と言う。行列Aの<math>(i,j)</math>成分が、<math>a_{ij}</math>である行列<math>A</math>を<math>A=(a_{ij})</math>と書く。 行列の第k列の列ベクトルを<math>\mathbf a_k = \begin{pmatrix} a_{1k} \\ a_{2k} \\ \vdots \\ a_{mk} \end{pmatrix} </math>とする。 行列は、この列ベクトルを用いて、行列は、<math>A = ( \mathbf a_1, \mathbf a_2, \cdots, \mathbf a_n)</math>と表すこともできる。 同じように、行列の第k行の行ベクトルを<math>\mathbf a_k = ( a_{k1}, a_{k2}, \cdots, a_{km})</math>としたとき。 行列は、この行ベクトルを用いて、行列は、<math>A = \begin{pmatrix} \mathbf a_1 \\ \mathbf a_2 \\ \vdots \\ \mathbf a_n \end{pmatrix}</math>と表すこともできる。 成分が全て実数の行列を'''実行列'''と言い、成分が全て複素数の行列を'''複素行列'''という。また、<math>m=n</math>の場合、(n×n)-行列を特に'''n次正方行列'''と呼ぶ。 === 相等関係 === 2つの(m×n)-行列<math>A,B</math>に関し、<math>A</math>と<math>B</math>が等しいとは、2つの行列の対応する成分が全て等しいことを言う。すなわち、 :<math>A=(a_{ij}), B=(b_{ij})</math>のとき、 :<math>A = B \iff \forall i,j , a_{ij} = b_{ij}</math> ===加法=== 2個のm行n列行列<math>A</math>と<math>B</math>について、行列の和 A+B を次のように定義する。 <math>A=\begin{pmatrix} a_{11} & a_{12} & \cdots & a_{1n}\\ a_{21} & a_{22} & \cdots & a_{2n}\\ \vdots & \vdots & \ddots & \vdots \\ a_{m1} & a_{m2} & \cdots & a_{mn}\\ \end{pmatrix}</math>, <math>B=\begin{pmatrix} b_{11} & b_{12} & \cdots & b_{1n}\\ b_{21} & b_{22} & \cdots & b_{2n}\\ \vdots & \vdots & \ddots & \vdots \\ b_{m1} & b_{m2} & \cdots & b_{mn}\\ \end{pmatrix}</math> のとき、 :<math>A+B=\begin{pmatrix} a_{11}+b_{11} & a_{12}+b_{12} & \cdots & a_{1n}+b_{1n}\\ a_{21}+b_{21} & a_{22}+b_{22} & \cdots & a_{2n}+b_{2n}\\ \vdots & \vdots & \ddots & \vdots\\ a_{m1}+b_{m1} & a_{m2}+b_{m2} & \cdots & a_{mn}+b_{mn}\\ \end{pmatrix}</math> <math>A=(a_{ij}),B=(b_{ij})</math>のとき、<math>A+B=(a_{ij} + b_{ij})</math>と表現することもできる。 ===スカラー乗法=== また、行列<math>A</math>と定数<math>\lambda</math>について、行列の定数倍 <math>\lambda A</math> を次のように定義する。 <math>A=\begin{pmatrix} a_{11} & a_{12} & \cdots & a_{1n}\\ a_{21} & a_{22} & \cdots & a_{2n}\\ \vdots & \vdots & \ddots & \vdots \\ a_{m1} & a_{m2} & \cdots & a_{mn}\\ \end{pmatrix}</math> のとき、 :<math>\lambda A=\begin{pmatrix} \lambda a_{11} & \lambda a_{12} & \cdots & \lambda a_{1n}\\ \lambda a_{21} & \lambda a_{22} & \cdots & \lambda a_{2n}\\ \vdots & \vdots & \ddots & \vdots \\ \lambda a_{m1} & \lambda a_{m2} & \cdots & \lambda a_{mn}\\ \end{pmatrix}</math> 特に、<math>\lambda=-1</math>のとき、<math>(-1)A</math>を<math>-A</math>と書く。 また、<math>A+(-B)</math>を<math>A-B</math>と書く。 === 積 === [[File:Matrix multiplication row column correspondance.svg|right|300px|300px|thumb|行列の積の計算過程の図示。行列''A''の第''i'' 行と行列''B''の第''j'' 列の各成分の積を実線部分のように取り、続いて点線のように加えていくことにより、積''AB''の''ij''成分を得る。]] [[File:Matrix multiplication qtl1.svg|thumb|行列の積では、行列''A''の列数と行列''B''の行数が同じでなくてはいけない。行列''C''の行数は行列''A''の行数であり、行列''C''の列数は行列''B''の列数である。]] 2個の行列<math>A</math>と<math>B</math>について、Aの列数とBの行数が同じで<math>A, B</math>の場合に、行列の積<math>AB</math>を次のように定義する。 <math>A=\begin{pmatrix} a_{11} & a_{12} & \cdots & a_{1n}\\ a_{21} & a_{22} & \cdots & a_{2n}\\ \vdots & \vdots & \ddots & \vdots\\ a_{m1} & a_{m2} & \cdots & a_{mn}\\ \end{pmatrix}</math> , <math>B=\begin{pmatrix} b_{11} & b_{12} & \cdots & b_{1p}\\ b_{21} & b_{22} & \cdots & b_{2p}\\ \vdots & \vdots & \ddots & \vdots\\ b_{n1} & b_{n2} & \cdots & b_{np}\\ \end{pmatrix}</math> のとき、AとBの積ABを : <math>c_{ij}=\sum_{k=1}^n a_{ik}b_{kj} = a_{i1}b_{1j} + a_{i2}b_{2j} + \cdots + a_{in}b_{nj}~(i=1,\cdots,m,~j=1,\cdots,p)</math> によって : <math>AB=\begin{pmatrix} c_{11} & c_{12} & \cdots & c_{1p}\\ c_{21} & c_{22} & \cdots & c_{2p}\\ \vdots & \vdots & \ddots & \vdots\\ c_{m1} & c_{m2} & \cdots & c_{mp}\\ \end{pmatrix}</math> と定める。 この定義は難しく見えるが、行列<math>A</math>の<math>i</math>行目の行ベクトルと、行列<math>B</math>の<math>j</math>列目の列ベクトルの内積が行列<math>AB</math>の<math>(i,j)</math>成分になっているだけである。 行列同士の積は全ての二行列に対して定義されているわけではない。(m×n)-行列と(n×l)-行列の間にのみ定義されているのである。 '''例題''' 次の計算をせよ。 :<math>\begin{pmatrix} -7 & -6 & -5\\ -4 & -3 & -2\\ \end{pmatrix} \begin{pmatrix} -1 & 0 &1\\ 2 & 3 & 4\\ 5 & 6 & 7\\ \end{pmatrix}</math> ;解答 :<math>\begin{align} \begin{pmatrix} -7 & -6 & -5\\ -4 & -3 & -2\\ \end{pmatrix} \begin{pmatrix} -1 & 0 &1\\ 2 & 3 & 4\\ 5 & 6 & 7\\ \end{pmatrix} &= \begin{pmatrix} (-7)(-1)+(-6)\times 2+(-5)\times 5 & (-7)\times 0+(-6)\times 3+(-5)\times 6 & (-7)\times 1+(-6)\times 4+(-5)\times 7\\ (-4)(-1)+(-3)\times 2+(-2)\times 5 & (-4)\times 0+(-3)\times 3+(-2)\times 6 & (-4)\times 1+(-3)\times 4+(-2)\times 7\\ \end{pmatrix} \\ &=\begin{pmatrix} -30 & -48 & -66\\ -12 & -21 & -30\\ \end{pmatrix} \end{align}</math> '''例題''' <math>m\times n</math> 行列 <math>A</math>、<math>n\times p</math> 行列 <math>B</math>、<math>p\times q</math> 行列 <math>C</math> について、 <math>A(BC)=(AB)C</math> を証明せよ。 ;解答 ここでは、行列 <math>X</math> の <math>(i,j)</math> 成分を <math>[X]_{ij}</math> と表すことにする。 <math>\begin{align} {[A(BC)]_{ij}} &= \sum_{k=1}^n[A]_{ik}[BC]_{kj} \\ &= \sum_{k=1}^n\left([A]_{ik} \sum_{l=1}^p[B]_{kl}[C]_{lj}\right)\\ &= \sum_{k=1}^n\sum_{l=1}^p[A]_{ik}[B]_{kl}[C]_{lj} \\ &= \sum_{l=1}^p\left([C]_{lj}\sum_{k=1}^n[A]_{ik}[B]_{kl}\right)\\ &= \sum_{l=1}^p [AB]_{il}[C]_{lj}\\ &= [(AB)C]_{ij} \end{align} </math> よって、<math>(AB)C=A(BC)</math> ===零行列=== 行列成分が全て0の行列を'''零行列'''(zero matrix)といい、<math>0</math>と書く。特に(m×n)-行列であることを明示する場合には、0{{sub|m,n}}と書き、n次正方行列であることを明示する場合には0{{sub|n}}と書く。 任意の行列に、適当な零行列をかけると、常に零行列が得られる。零行列は、実数における0に似ている。 ===単位行列=== <math>A=(a_{ij})</math> に対して、成分<math>a_{kk}, ~k=1,\cdots,n</math>を、<math>n</math>次正方行列<math>A</math>の'''対角成分'''(diagonal element)という。 行列の対角成分がすべて1で、その他の成分がすべて0であるような正方行列 <math>I =\begin{pmatrix} 1 & 0 & \cdots & 0 \\ 0 & 1 & \cdots & 0 \\ \vdots & \vdots & \ddots & \vdots \\ 0 & 0 & \cdots & 1 \end{pmatrix}</math>を'''単位行列'''(elementary matrix、あるいはidentity matrix)といい、<math>E_n</math>や<math>I_n</math>と表す。<math>n</math>が明らかである場合にはしばしば省略して、<math>E</math>や<math>I</math>と表すこともある。クロネッカーのデルタを使うと<math>I = (\delta _{ij})</math>. ==行列の演算の性質== <math>A,B,C</math>を任意の<math>(m,n)</math>行列 、<math>\lambda, \mu</math>を任意の定数、<math>0</math>を零行列、<math>I</math>を単位行列とすると、以下の関係が成り立つ。 #結合法則: <math>(A+B)+C=A+(B+C)</math> #交換法則: <math>A+B=B+A</math> #<math>\lambda (A+B)=\lambda A+ \lambda B</math> #<math>(\lambda +\mu )A=\lambda A+\mu A</math> #<math>(\lambda \mu )A=\lambda (\mu A)</math> #<math>1A=A</math> #<math>0A=0</math> #<math>A+0=A</math> #<math>A-A=O</math> #<math>A(B+C)=AB+AC</math> #<math>(A+B)C=AC+BC</math> #<math>A0 = 0A = 0</math> #<math>AI = IA = A</math> == 転置行列 == [[画像:Matrix transpose.gif|200px|right]] <math>A=\begin{pmatrix} a_{11} & a_{12} & \cdots & a_{1n}\\ a_{21} & a_{22} & \cdots & a_{2n}\\ \vdots & \vdots & \ddots & \vdots\\ a_{m1} & a_{m2} & \cdots & a_{mn} \end{pmatrix}</math> に対して :<math>A^\mathsf T = \begin{pmatrix} a_{11} & a_{21} & \cdots & a_{m1}\\ a_{12} & a_{22} & \cdots & a_{m2}\\ \vdots & \vdots & \ddots & \vdots\\ a_{1n} & a_{2n} & \cdots & a_{mn}\\ \end{pmatrix}</math> を<math>A</math>の'''転置行列'''(transposed matrix)と言い、<math>A^\mathsf T </math>や<math>^tA</math>と表す。 つまり<math>A^\mathsf T</math>とは、<math>A</math>の縦横をひっくり返した行列である。 以下のような性質が成り立つ。 # <math>(A^\mathsf T)^\mathsf T = A</math> # <math>(A+B)^\mathsf T = A^\mathsf T + B^\mathsf T </math> # <math>(\lambda A)^\mathsf T = \lambda (A^\mathsf T)</math> # <math>(AB)^\mathsf T = B^\mathsf T A^\mathsf T </math> ;証明 <math>A=(a_{ij}),B=(b_{ij})</math>とする。 #転置行列とは、行と列を入れ替えた行列なので、2回行と列を入れ替えれば、もとの行列に戻る。 #<math>(A+B)^\mathsf T</math>の<math>(i,j)</math>成分は<math>a_{ji}+b_{ji}</math>であり、<math>A ^\mathsf T +B ^\mathsf T </math>の<math>(i,j)</math>成分は<math>a_{ji}+b_{ji}</math>である。 #<math>(\lambda A)^\mathsf T </math>の<math>(i,j)</math>成分は<math> \lambda a_{ji}</math>であり、<math>\lambda(A ^\mathsf T) </math>の<math>(i,j)</math>成分は<math> \lambda a_{ji}</math>であるから。 #<math>AB </math>の<math>(i,j)</math>成分は<math> = \sum_{k=1}^{n} a_{ik} b_{kj}</math>なので、<math>(AB) ^\mathsf T </math>の<math>(i,j)</math>成分は<math> \sum_{k=1}^{n} a_{jk} b_{ki}</math>である。次に、<math>A ^\mathsf T </math>の<math>(i,j)</math>成分は<math> a_{ji}, B ^\mathsf T </math>の<math>(i,j)</math>成分は<math> b_{ji}</math>であるので、<math>A ^\mathsf T B ^\mathsf T </math>の<math>(i,j)</math>成分は<math> \sum_{k=1}^{n} a_{jk} b_{ki}</math>であるから。 ただし、<math>n</math>を<math>A</math>の列数とする。 == 複素行列 == ある行列Aのすべての成分の複素共役を取った行列<math>\overline A= \begin{pmatrix} \overline a_{1,1} & \cdots & \overline a_{1,m} \\ \vdots & \ddots & \vdots \\ \overline a_{n,1} & \cdots & \overline a_{m,n} \end{pmatrix}</math> を、'''複素共役行列'''(complex conjugate matrix)という。 以下のような性質がある。 * <math>\overline{\overline{A}}=A</math> * <math>\overline{A+B}=\overline{A}+\overline{B}</math> * <math>\overline{\lambda A}=\overline{\lambda}~\overline{A}</math> * <math>\overline{AC}=\overline{C}~\overline{A}</math> 一番最後の式には注意せよ。とりあえず、ここで一休みして、演習をやろう。 '''演習''' :1.定理(1.5.1)を証明せよ :2.計算せよ ::(1)<math>\begin{pmatrix} 1 & -2 & 3\\ 2 & 1 & 0\\ \end{pmatrix} \begin{pmatrix} -2 & -1 & 1\\ 1 & 2 & -3\\ 2 & 0 & 3\\ \end{pmatrix}</math> ::(2)<math>\begin{pmatrix} 1 & 1 & 1\\ 1 & 1 & 1\\ 1 & 1 & 1\\ \end{pmatrix} \begin{pmatrix} 2\\ 3\\ 4\\ \end{pmatrix}</math> ::(3)<math>\begin{pmatrix} 1 & i\\ -i & 1\\ \end{pmatrix} \begin{pmatrix} 1-i & 0\\ 0 & 1+i\\ \end{pmatrix}</math> ::(4)<math>\begin{pmatrix} 2 & 1+i & 3i\\ \end{pmatrix} \begin{pmatrix} 2\\ 1+i\\ -3i\\ \end{pmatrix}</math> :(<math>i=\sqrt{-1}</math>) 3. :対角成分*<sup>1</sup>が全て1それ以外の成分が全て0のn次正方行列*<sup>2</sup>を、単位行列と言い、E<sub>n</sub>と書く。つまり、 :<math>E_n=(\delta_{i,j})</math>, <math>\delta_{i,j}=\begin{cases} 1(i=j)\\ 0(i\neq j) \end{cases}</math> :このδ<sub>i,j</sub>を、クロネッカーのデルタ(Kronecker delta)と言う、またはクロネッカーの記号と言う。この時、次のことを示せ。 ::(1)<math>A=\begin{pmatrix} -1 & -2\\ -2 & -4\\ \end{pmatrix}</math>のとき、AX=E<sub>2</sub>を満たすXは存在しない ::(2)<math>B=\begin{pmatrix} -1 & -2\\ 0 & 0\\ \end{pmatrix}</math>の時、(1)の定義で、BX=AとなるXが存在しない。 ::また、YB=Aを満たすYが無数に存在する。 ::(3)n次行列(n次正方行列)Aのある列が全て0なら、AX=Eを満たすXは存在しない。 :<nowiki>*</nowiki><sup>1</sup>対角成分:n次正方行列A=(a<sub>i,j</sub>)で、(i=1,2,...,n;j=1,2,...,n)a<sub>i,i</sub>=a<sub>1,1</sub>,a<sub>2,2</sub>,...,a<sub>n,n</sub>のこと :<nowiki>*</nowiki><sup>2</sup>n次正方行列:行と、列の数が同じnの時の行列 ==区分け== <math>A=\begin{pmatrix} 1 & 3 & -1 & 0\\ 2 & 1 & 3 & 2\\ 1 & 2 & 3 & 4\\ \end{pmatrix}</math>は、 <math>A_1=\begin{pmatrix} 1 & 3 & -1\\ 2 & 1 & 3\\ \end{pmatrix}</math>, <math>A_2=\begin{pmatrix} 0\\ 2\\ \end{pmatrix}</math> <math>A_3=\begin{pmatrix} 1 & 2 & 3\\ \end{pmatrix}</math>, <math>A_4=\begin{pmatrix} 4\\ \end{pmatrix}</math> とすることで、 <math>A=\begin{pmatrix} A_1 & A_2\\ A_3 & A_4\\ \end{pmatrix}</math> 一般に、 定義(2.1)行列の区分け (l,m)型行列A=(a<sub>i,j</sub>)をp-1本の横線とq-1本の縦線でp×qの島に分けて、上からs番目、左からt番目の行列をA<sub>s,t</sub>とおいて、 <math>A=\begin{pmatrix} A_{1,1} & A_{1,2} & A_{1,3} & \cdots & A_{1,q}\\ A_{2,1} & A_{2,2} & A_{2,3} & \cdots & A_{2,q}\\ A_{3,1} & A_{3,2} & A_{3,3} & \cdots & A_{3,q}\\ \vdots & \vdots & \vdots & \ddots & \vdots\\ A_{p,1} & A_{p,2} & A_{p,3} & \cdots & A_{p,q}\\ \end{pmatrix}</math> とすることを、行列の''区分け''と言う。 定理(2.2) 同様に区画された同じ型の、 <math>A=\begin{pmatrix} A_{1,1} & A_{1,2} & A_{1,3} & \cdots & A_{1,q}\\ A_{2,1} & A_{2,2} & A_{2,3} & \cdots & A_{2,q}\\ A_{3,1} & A_{3,2} & A_{3,3} & \cdots & A_{3,q}\\ \vdots & \vdots & \vdots & \ddots & \cdots\\ A_{p,1} & A_{p,2} & A_{p,3} & \cdots & A_{p,q}\\ \end{pmatrix}</math>, <math>B=\begin{pmatrix} B_{1,1} & B_{1,2} & B_{1,3} & \cdots & B_{1,r}\\ B_{2,1} & B_{2,2} & B_{2,3} & \cdots & B_{2,r}\\ B_{3,1} & B_{3,2} & B_{3,3} & \cdots & B_{3,r}\\ \vdots & \vdots & \vdots & \ddots & \vdots\\ B_{q,1} & B_{q,2} & B_{q,3} & \cdots & B_{q,r}\\ \end{pmatrix}</math> <math>C=AB=\begin{pmatrix} C_{1,1} & C_{1,2} & C_{1,3} & \cdots & C_{1,r}\\ C_{2,1} & C_{2,2} & C_{2,3} & \cdots & C_{2,r}\\ C_{3,1} & C_{3,2} & C_{3,3} & \cdots & C_{3,r}\\ \vdots & \vdots & \vdots & \ddots & \vdots\\ C_{p,1} & C_{p,2} & C_{p,3} & \cdots & C_{p,r}\\ \end{pmatrix}</math> がある。この時、 <math>C_{s,u}=\sum_{t=1}^q A_{s,t}B_{t,u}=A_{s,1}B_{1,u}+A_{s,2}B_{2,u}+...+A_{s,q}B_{q,u}</math> (2.3) (s=1,2,...,p;u=1,2,...,r) (証明) :(i) ::A<sub>s,t</sub>を(l<sub>s</sub>,m<sub>t</sub>),B<sub>t,u</sub>を(m<sub>t</sub>,n<sub>u</sub>)とすると、A<sub>s,t</sub>B<sub>t,u</sub>は、tと関係なく、(l<sub>s</sub>,m<sub>t</sub>)型行列であるから、それらの和C<sub>s,u</sub>も(l<sub>s</sub>,m<sub>t</sub>)型行列である。よって、(2.3)は意味を成す。 :(ii) ::Aを(l,m)Bを(m,n)型、(2.3)の両辺の対応する成分を(α,β)、 ::<math>i=\sum_{j=1}^{s-1} l_j+\alpha=l_1+l_2+\dot+l_{s-1}+\alpha</math>, ::<math>k=\sum_{j=1}^{u-1} n_j+\beta=n_1+n_2+\dot+n_{u-1}+\beta</math>. ::とおけば、C<sub>s,u</sub>の(α,β)成分とCの(i,k)成分,A<sub>s,t</sub>B<sub>t,u</sub>は等しく、それは ::<math>\sum_{j=1}^m a_{i,j}b_{j,k}</math>であり且 ::<math>\sum_{j=(\sum_{j=1}^{t-1} m_j)+1}^{\sum_{j=1}^t m_t} a_{i,j}b_{j,k}</math> ::⇔<math>\sum_{t=1}^q A_{s,t}B{t,u}</math>の(α,β)成分=<math>\sum_{j=1}^m a_{i,j}b_{j,k}</math> :(i),(ii)より、定理(2.2)は証明された。 '''例''' :p=q=r=2とすると、<math>\begin{pmatrix} A_{1,1} & A_{1,2}\\ A_{2,1} & A_{2,2}\\ \end{pmatrix} \begin{pmatrix} B_{1,1} & B_{1,2}\\ B_{2,1} & B_{2,2}\\ \end{pmatrix}=\begin{pmatrix} A_{1,1}B_{1,1}+A_{1,2}B_{2,1} & A_{1,1}B_{1,2}+A_{1,2}B_{2,2}\\ A_{2,1}B_{1,1}+A_{2,2}B_{2,1} & A_{2,1}B_{1,2}+A_{2,2}B_{2,2}\\ \end{pmatrix}</math> (2.4) :A<sub>2,1</sub>,B<sub>2,1</sub>=Oとすると、(2.4)右辺は :<math>\begin{pmatrix} A_{1,1}B_{1,1} & A_{1,1}B_{1,2}+A_{1,2}B_{2,2}\\ O & +A_{2,2}B_{2,2}\\ \end{pmatrix}</math> :と、区分けはこの時威力を発揮する。A<sub>1,2</sub>,B<sub>1,2</sub>=Oならさらに威力を発揮する。 :<math>\begin{pmatrix} A_{1,1}B_{1,1} & O\\ O & +A_{2,2}B_{2,2}\\ \end{pmatrix}</math> '''例''' :単位行列E<sub>n</sub>をn個の縦ベクトルに分割したときの、そのベクトルをn項単位ベクトルと言う。これは、ベクトルの項でのべた、2,3次における単位ベクトルの定義の一般化である。Eのことを単位行列と言う意味が分かっただろうか。ここでAを、(l,m)型Bを(m,n)型と定義しなおし、 :B=('''b'''<sub>1</sub>,'''b'''<sub>2</sub>,...,'''b'''<sub>n</sub>) :とすると、 :AB=(A'''b'''<sub>1</sub>,A'''b'''<sub>2</sub>,...,A'''b'''<sub>n</sub>) :この事実は、定理(2.2)の特殊化である。 縦ベクトル'''x'''=(x<sub>i</sub>)は、 '''x'''=x<sub>1</sub>'''e'''<sub>1</sub>+x<sub>2</sub>'''e'''<sub>2</sub>+...+x<sub>k</sub>'''e'''<sub>k</sub> と表す事が出来るが、一般に x<sub>1</sub>'''a'''<sub>1</sub>+x<sub>2</sub>'''a'''<sub>2</sub>+...+x<sub>k</sub>'''a'''<sub>k</sub> を'''a'''<sub>1</sub>,'''a'''<sub>2</sub>,...,'''a'''<sub>k</sub>の''線型結合''と言う。 '''演習''' 計算せよ (1)<math>\begin{pmatrix} 1 & 1 & 0 & 0\\ 0 & -2 & 0 & 0\\ 0 & 0 & -2 & 3\\ 0 & 0 & 1 & 1\\ \end{pmatrix} \begin{pmatrix} 2 & 1 & 0 & 0\\ 0 & 1 & 0 & 0\\ 0 & 0 & 1 & 1\\ 0 & 0 & -2 & 3\\ \end{pmatrix}</math> (2)<math>\begin{pmatrix} 2i & 0 & 0\\ 0 & -i & 0\\ 0 & 0 & 3i\\ \end{pmatrix} \begin{pmatrix} -1 & 0 & 0\\ 0 & 2 & 0\\ 0 & 0 & 1\\ \end{pmatrix}(i=\sqrt{-1})</math> ==逆行列== <math>AX=XA=I</math>となる行列<math>X</math>が存在すれば、<math>X</math>を<math>A</math>の逆行列といい、<math>A^{-1}</math>と表す。<br> また、<math>A</math>に逆行列が存在すれば、<math>A</math>を'''正則行列'''といい、逆行列はただ一通りに決まる。<br> '''証明''' :<math>A</math>に逆行列<math>X,Y</math>が存在すると仮定すると。 :<math>AX=XA=I,AY=YA=I</math>が成り立つので、 :<math>X=XI=X(AY)=(XA)Y=IY=Y</math>よって<math>X=Y</math>となるので、逆行列が存在すれば、ただ一通りに決まる。 逆行列については、以下の性質が成り立つ。 #<math>(A^{-1})^{-1}=A</math> #<math>(AB)^{-1}=B^{-1} A^{-1}</math> '''証明''' :#<math>A^{-1}</math>の逆行列は、定義から、<math>A^{-1}X=X A^{-1} = I</math>となる<math>X</math>であるが、<math>X</math>に<math>A</math>を代入すると成り立っているので、<math>(A^{-1})^{-1}=A</math>である。 :#<math>AB</math>の逆行列は、<math>(AB)X=X(AB)=I</math>となる<math>X</math>であるが、<math>X</math>に<math>B^{-1} A^{-1}</math>を代入すると、 :::<math>(AB)(B^{-1} A^{-1})=AI A^{-1} = A A^{-1} = I</math> :::<math>(B^{-1} A^{-1})(AB)=B^{-1} I B =B^{-1} B = I </math> ::となり、式が成り立っているので<math>(AB)^{-1}=B^{-1} A^{-1}</math>である。 定義(3.2.4)対称区分け 正方行列を一辺が等しい正方形の島に区分けするとき、この区分けを''対称区分け''と言う。 簡単な証明で 「定理(3.2.5) 対称区分けで、 <math>A=\begin{pmatrix} A_{1,1} & A_{1,2}\\ O & A_{2,2}\\ \end{pmatrix}</math> において、A<sub>1,1</sub>とA<sub>2,2</sub>が正則ならば、Aも正則である。」 及び次のことが言える。 「対称区分けで、 A=(A<sub>i,j</sub>)で、(i,j=1,2,...n)<math>A_{i,j}=\begin{cases} A_i(i=j)\\ O \end{cases}</math>ならば、Aが正則である必要十分条件は、A<sub>i</sub>がすべて正則である事である」 その逆行列は、次のように与えられる。 <math>A^{-1}=\begin{cases} A_i^{-1}(i=j)\\ O \end{cases}</math> また、(3.2.5)の逆行列A<sub>-1</sub>は、 <math>A^{-1}=\begin{pmatrix} A_{1,1}^{-1} & -A_{1,1}^{-1}A_{1,2}A^{-1}_{2,2}\\ O & A_{2,2}^{-1} \end{pmatrix}</math>である。 ==行列の累乗== 行列の累乗は、<math>A</math>を正則行列、<math>n</math>を自然数とし、次のように定義される。<br> :<math>A^n=\underbrace{AA...A}_n</math><br> :<math>A^0=I</math><br> :<math>A^{-n}=\underbrace{A^{-1}A^{-1}...A^{-1}}_n</math><br> 行列の累乗には以下の性質がある。<br> :<math>A^k A^l = A^{k + l}</math><br> :<math>(A^k)^l = A^{kl}</math><br> :<math>AB=BA</math>のとき<math>(AB)^k=A^k B^k</math> ただし:<math>A</math>を正則行列、<math>k,l</math>を自然数とする。 '''証明'''<br><math>(AB)^k = ABABAB \cdots ABAB</math> :<math>AB=BA</math>なので、隣り合うAとBを入れ替えていくと<br> <math>ABABAB \cdots ABAB=AABABA \cdots BABB</math><br> これを続けると、<math>\underbrace{AA \cdots A}_k \underbrace{BB \cdots B}_k</math>となる。 ==その他== 正方行列(a<sub>i,j</sub>)において、a<sub>i,i</sub>を対角成分と言う。また、対角成分以外が全て0である正方行列のことを''対角行列''(diagonal matrix)と言う。対角行列が正則であるための、必要十分条件は、対角成分が全て0でないということである。4章で示される。対角行列の中でも更にスカラー行列と呼ばれるものがある。それはcE(c≠0)の事である。勿論Eはc=1の時のスカラー行列で、対角行列である。また、スカラー行列cEを任意行列Aに掛けると、CAとでる。対角行列が定義されたので、固有和が定義できる。 定義(3.2.6)固有和または跡(trace) :正方行列Aの固有和 :TrA :とは、対角成分の総和である。 次のような性質がある Tr(cA)=cTrA, Tr(A+B)=TrA+TrB, Tr(AB)=Tr(BA) {{ナビゲーション|本=[[線型代数学]]|前ページ=[[線型代数学/ベクトル|ベクトル]]|ページ名=行列概論|次ページ=[[線型代数学/線型方程式|線型方程式]]}} [[Category:線形代数学|きようれつかいろん]] [[カテゴリ:行列]]
このページで使用されているテンプレート:
テンプレート:Sub
(
ソースを閲覧
)
テンプレート:ナビゲーション
(
ソースを閲覧
)
線型代数学/行列概論
に戻る。
ナビゲーション メニュー
個人用ツール
ログイン
名前空間
ページ
議論
日本語
表示
閲覧
ソースを閲覧
履歴表示
その他
検索
案内
メインページ
最近の更新
おまかせ表示
MediaWiki についてのヘルプ
特別ページ
ツール
リンク元
関連ページの更新状況
ページ情報