1. 用图表检验Analyze -> regression -> Linear-> Plots
1 n: i& b: t" m% u- DScatter plot of the standardised residuals on the standardised predicted values (ZRESID as the Y variable, and ZPRED as the X variable1 h4 @6 X% p; {; p" w3 Q+ R7 x
+ J% X. {3 t% s6 [$ O# u如果图表显示有可能存在异方差,需要用统计检验来进一步检测异方差是否确实存在。
: A% B' ` B1 t( V! m/ g2 F2. 用统计检验
0 ?! E2 s% O! Q3 ~8 I: n1 U2 O( sHeteroscedasticity——Testing and Correcting in SPSS.pdf
Gwilym Pryce March 2002.doc
(172.5 KB, 下载次数: 4)
: _* b' Y; s) R, l" @$ _% L \: |, F) w! a& @# R# L3 L
' |; `. d) K- o6 w3 v
Levene’s Test
4 _ n- k5 C- b1 W/ RGoldfeld-Quandt Test3 ]4 x1 [. Y, l. `5 s
Breusch-Pagan Test
6 ?! N- h" C, R" |3 f& w( yWhite‘s Test (比较常用来检验异方差)
5 H0 N/ d% v2 H% MAssume you want to run a regression of wage on age, work experience,education, gender, and a dummy for sectorofemployment (whether employed in the public sector). wage = function(age, workexperience, education, gender, sector) or, as your textbook will have it, wage = b1 + b2*age + b3*work experience+ b4*education + b5*gender + b6*sector The White’s test is usually used as a test for heteroskedasticity. In this test, a regression of the squares ofthe residuals is run on the variables suspected of causing theheteroskedasticity, their squares, and cross products. (residuals)2 = b0 + b1 educ + b2 work_ex+ b3 (educ)2 + b4 (work_ex)2 + b5(educ*work_ex)
1 P* f- Z' \5 A- Q) U0 {/ y" q' F3 r' |/ A
! ?1 |* Z5 @' |: _) F' @
8 Q& r' M, x+ k2 n White’s Test · Calculate n*R2 à R2 = 0.037, n=2016 à Thus, n*R2 = .037*2016 = 74.6. : B8 {( t, E: g: z
· Compare this value with c2 (n), i.e.with c2 (2016) & d0 i$ a5 H( m9 t6 o9 \0 W
(c2 is the symbol for theChi-Square distribution) 7 _ H2 t4 h( p h
c2 (2016) = 124obtained from c2 table. (For 955 confidence) As n*R2 < c2 ,heteroskedasticity can not be confirmed. ! k* A$ b) b/ K
; S" U9 o; Y r/ C# d' S
请参考:regression_explained_SPSS
regression_explained_SPSS.doc
(368 KB, 下载次数: 0)
7 ^# L0 h4 x" F' H. N! |
|