如果有n次样本观测值,则:
- Y i = β 1 ^ + β 2 ^ X 2 i + β 3 ^ X 3 i + . . . + β k ^ X k i + e i Y_i=\hat{β_1}+\hat{β_2}X_{2i}+\hat{β_3}X_{3i}+…+\hat{β_k}X_{ki}+e_i Yi=β1+β2X2i+β3X3i+…+βkXki+ei
对被解释变量Y和多个解释变量做n次观测,所得的n次观测值可写为方程组形式:
Y 1 = β 1 + β 2 X 21 + β 3 X 31 + . . . + β k X k 1 + u 1 Y_1=β_1+β_2X_{21}+β_3X_{31}+…+β_kX_{k1}+u_1 Y1=β1+β2X21+β3X31+…+βkXk1+u1
Y 2 = β 1 + β 2 X 22 + β 3 X 32 + . . . + β k X k 2 + u 2 Y_2=β_1+β_2X_{22}+β_3X_{32}+…+β_kX_{k2}+u_2 Y2=β1+β2X22+β3X32+…+βkXk2+u2
…
Y n = β 1 + β 2 X 2 n + β 3 X 3 n + . . . + β k X k n + u n Y_n=β_1+β_2X_{2n}+β_3X_{3n}+…+β_kX_{kn}+u_n Yn=β1+β2X2n+β3X3n+…+βkXkn+un
该方程组也可表示为矩阵形式:
[ Y 1 Y 2 . . . Y n ] = [ 1 X 21 X 31 . . . X k 1 1 X 22 X 32 . . . X k 2 . . . . . . . . . . . . . . . 1 X 2 n X n . . . X k n ] [ β 1 β 2 . . . β k ] + [ u 1 u 2 . . . u n ] \begin{gathered} \begin{bmatrix}Y_1 \\ Y_2 \\ … \\ Y_n \end{bmatrix}= \begin{bmatrix} 1 & X_{21} & X_{31}& … & X_{k1} \\ 1 & X_{22} & X_{32}& … & X_{k2} \\ … & … & … & … & … \\ 1 & X_{2n} & X_{n}& … & X_{kn} \end{bmatrix} \begin{bmatrix} β_1 \\ β_2 \\ … \\ β_k \end{bmatrix}+ \begin{bmatrix} u_1 \\ u_2 \\ … \\ u_n \end{bmatrix} \end{gathered} ⎣⎢⎢⎡Y1Y2…Yn⎦⎥⎥⎤=⎣⎢⎢⎡11…1X21X22…X2nX31X32…Xn…Xk1Xk2…Xkn⎦⎥⎥⎤⎣⎢⎢⎡β1β2…βk⎦⎥⎥⎤+⎣⎢⎢⎡u1u2…un⎦⎥⎥⎤
记
Y = [ Y 1 Y 2 . . . Y n ] β = [ β 1 β 2 . . . β k ] U = [ u 1 u 2 . . . u n ] \begin{gathered} Y= \begin{bmatrix}Y_1 \\ Y_2 \\ … \\ Y_n \end{bmatrix} β= \begin{bmatrix} β_1 \\ β_2 \\ … \\ β_k \end{bmatrix} U= \begin{bmatrix} u_1 \\ u_2 \\ … \\ u_n \end{bmatrix} \end{gathered} Y=⎣⎢⎢⎡Y1Y2…Yn⎦⎥⎥⎤β=⎣⎢⎢⎡β1β2…βk⎦⎥⎥⎤U=⎣⎢⎢⎡u1u2…un⎦⎥⎥⎤
X = [ 1 X 21 X 31 . . . X k 1 1 X 22 X 32 . . . X k 2 . . . . . . . . . . . . . . . 1 X 2 n X n . . . X k n ] X= \begin{gathered} \begin{bmatrix} 1 & X_{21} & X_{31}& … & X_{k1} \\ 1 & X_{22} & X_{32}& … & X_{k2} \\ … & … & … & … & … \\ 1 & X_{2n} & X_{n}& … & X_{kn} \end{bmatrix} \end{gathered} X=⎣⎢⎢⎡11…1X21X22…X2nX31X32…Xn…Xk1Xk2…Xkn⎦⎥⎥⎤
这里的X是由解释变量 X i j X_{ij} Xij的数据构成的矩阵,其中截距项可视为解释变量总是取值为1。X一般是由非随机变量构成,有时也称为X的数据矩阵或设计矩阵。
这样多元总体线性回归函数的矩阵形式可表示为
- Y = X β + U Y=Xβ+U Y=Xβ+U
E