x y x y x y
x y x y yfxc1c2.cm 0-0-1 c1c2.cm m xiyii12.N xy m 0-0-1 yifxc1c2.cm 0-0-2 i12.m. m m N<m N>m 0-0-2 m y yi <fxc1c2.cm> yi
p( yi ) =
[ y i - f ( xi ; c1 , c 2 ,., c m ) exp- 2 i2 2 i 1
]
2
,
i C c c .c y y .c 1 2 m 1 2 N
L=
(
i
1 N [ y
OLS
1
BLUE t F
2
1 Y=f(X1,X2,.,XP) Y Xi i=1,2.p 2 : Y=f(X1,X2,.,XP) Y Xi i=1,2.p
3
21 GDP
4
2-1 M2 y GDP x 1995 2004
5
x y x y x y
6
" regressive relationship , regression equation regression model
7
2-1
y= + x
2.1
x y 2.1 y y 21
8
Gauss and the Method of Least Squares
Teddy Petrou Hongxiao Zhu
1
Outline
Who was Gauss? Why was there controversy in finding the method of least squares? Gauss' treatment of error Gauss' derivation of the method of least squares Gauss' derivation by