Page 103 - Basic Statistics

98

5.2 THE MULTIPLE LINEAR REGRESSION MODEL

5.2.1 MODEL AND ESTIMATION COEFFICIENTS

Regresi ganda memodelkan hubungan antara suatu variabel terikat ( Y )

dengan beberapa variabel bebas ( Xi ). Model aditip linier bagi regresi ganda

adalah:
Multiple regression models is a model of the relationship between a dependent

variable (Y) with several independent variables (Xi). Linear additive model for

multiple regression is:

Y j = 0 + 1X1 j + 2 X2 j + …+ p Xp j +  j (G.1)

In the matrix and vector algebra expression,

Y = X +  (G.2)

or

 Y 1   1 X 1 1 X 2 1 X 3 1 . . . X p 1   β   ε 1 
     0   
 Y 2   1 X 1 2 X 2 2 X 3 2 . . . X p 2   β   ε 2 

     1   
 Y 3   1 X 1 3 X 2 3 X 3 3 . . . X p 3   β 2   ε 3 




.  = . . . . .   .  + . 


     
 .   . . . . .   .   . 
       
 .   . . . . .   .   . 
 Y n   1 X n 1 X n 2 X n 3 . . . X n p   β p   ε n 


       

( n x 1 ) ( n x (p+1) ) ( n x 1) ( n x 1 )

In association with regression modeling, the dependent variable Y is
often called the response variable, and the independent variable Xi is called
explanatory variables or regressors. Parameters  called regression coefficients,

while the difference between the expected value of Y in the model ( E(Y)= X )

with the actual value of Y, which is  called the error.

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