Page 123 - Basic Statistics

118

the other variables. While on sequeltial test, the coefficient of independent

variables were tested in order to enter. For example, the coefficient 3 was
tested when the regressor X3 is added to a model that involving only X1 , X2, and

the constant term.

Table 5.14 The sequential F-tests for linear regression model

Dependent Variable: Y
Source Type I Sum of df Mean Square F Sig.
Squares

a
Corrected Model 208,007 5 41,601 84,070 ,000
Intercept 387,157 1 387,157 782,379 ,000
X1 199,145 1 199,145 402,440 ,000
X2 ,127 1 ,127 ,256 ,620
X3 4,120 1 4,120 8,325 ,011
X4 ,263 1 ,263 ,532 ,476
X5 4,352 1 4,352 8,795 ,009

Error 7,918 16 ,495

Total 603,081 22

Corrected Total 215,925 21
a. R Squared = ,963 (Adjusted R Squared = ,952)

Furthermore, suppose we want to use the sequential sum of squares to
test on subsets of variables such as H0: 4 = 5 = 0. Dengan menggunakan hasil

pada tabel 5, The proper sum of squares is found by computing SS( 5 ,4  0 , 1

, 2, 3 ) and using the results in Table 5.12.

SS( 5 ,4  0 , 1 , 2, 3 ) = SS( 4  0 , 1 , 2 , 3 )+ SS( 5  0 , 1 , 2 , 3, 4 )

= 0.263 + 4.352

= 4.615

. 4 615 2 /
F = = 64 . 663
. 0 495

The numerator and denominator degrees of freedom are 2 and 16 respectively.

The hypothesis is rejected at the 0.025 level.

~~* CHAPTER 5 THE MULTIPLE LINEAR REGRESSION MODEL *~~