and one dependent (explanatory) variable and if we suppose a linear dependency,
we work with a simple linear regression with the following definition:
y ¼ b 1 þ b 2 :x þ E
ð5:1Þ
where y is dependent variable
b 1 is a constant value for independent variable equal to zero
b 2 is a slope of the line of dependency or a coefficient of the x 2 variable
x is independent/explanatory variable
E is a “noise” factor, which shows an influence of other phenomena, which are not
included in independent variables.
Multiple linear regression uses line of dependency and uses more than one
independent/explanatory values (x 1 , . . .. x n ), as the following equation shows
y ¼ b 1 þ b 2 :x 1 þ b 2 :x 2 þ þb 3 :x 3 þ . . . : b n :x n þ E
ð5:2Þ
where y is dependent variable
b 1 is a constant value for independent variable equal to zero
b 2 , b 3 , .. b n are coefficient of the x 2 , x 3, .. x n variables
Table 5.5 Results of multiple linear regression analysis for small cities before and after 1990
Indicator
Std. error
t- stat
P-value
Transport areas 70-90
21.59477837
À2.224422349 0.268961497
90-05
1.822978
À10.5839
0.059972
Residential areas 70-90
12.22371537
0.57080272
0.669801755
90-05
1.047294
14.81925
0.042894
Other areas 70-90
0.475742124
0.10104548
0.93589005
90-05
0.0432
8.582103
0.073847
Recreational areas 70-90
0.448298719
À2.679595562 0.227390043
90-05
0.036784
À26.6697
0.023859
Production areas 70-90
22.2153785
0.961241538 0.512579375
90-05
1.397767
10.05442
0.06311
Number of inhabitants 70-90
0.322577794
1.720817818 0.335129853
90-05
0.023357
7.200552
0.087851
Number of inhabitants/traffic area 70-90
9.1951904
À0.944071454 0.518309718
90-05
0.716079
À10.6351
0.059685
Number of inhabitants/production area
70-90
10.97723256
À0.478570031 0.715840227
90-05
0.796793
À6.45812
0.0978
Number of inhabitants/residential area
70-90
72.33747079
0.007553272 0.995191529
90-05
6.740035
7.310378
0.086547
R
2 value 70-90
0.97574884
R
2 value 90-05
0.99980001
5 Influence of Political Regime Change to Land Use Development in Urban. . .
101
we work with a simple linear regression with the following definition:
y ¼ b 1 þ b 2 :x þ E
ð5:1Þ
where y is dependent variable
b 1 is a constant value for independent variable equal to zero
b 2 is a slope of the line of dependency or a coefficient of the x 2 variable
x is independent/explanatory variable
E is a “noise” factor, which shows an influence of other phenomena, which are not
included in independent variables.
Multiple linear regression uses line of dependency and uses more than one
independent/explanatory values (x 1 , . . .. x n ), as the following equation shows
y ¼ b 1 þ b 2 :x 1 þ b 2 :x 2 þ þb 3 :x 3 þ . . . : b n :x n þ E
ð5:2Þ
where y is dependent variable
b 1 is a constant value for independent variable equal to zero
b 2 , b 3 , .. b n are coefficient of the x 2 , x 3, .. x n variables
Table 5.5 Results of multiple linear regression analysis for small cities before and after 1990
Indicator
Std. error
t- stat
P-value
Transport areas 70-90
21.59477837
À2.224422349 0.268961497
90-05
1.822978
À10.5839
0.059972
Residential areas 70-90
12.22371537
0.57080272
0.669801755
90-05
1.047294
14.81925
0.042894
Other areas 70-90
0.475742124
0.10104548
0.93589005
90-05
0.0432
8.582103
0.073847
Recreational areas 70-90
0.448298719
À2.679595562 0.227390043
90-05
0.036784
À26.6697
0.023859
Production areas 70-90
22.2153785
0.961241538 0.512579375
90-05
1.397767
10.05442
0.06311
Number of inhabitants 70-90
0.322577794
1.720817818 0.335129853
90-05
0.023357
7.200552
0.087851
Number of inhabitants/traffic area 70-90
9.1951904
À0.944071454 0.518309718
90-05
0.716079
À10.6351
0.059685
Number of inhabitants/production area
70-90
10.97723256
À0.478570031 0.715840227
90-05
0.796793
À6.45812
0.0978
Number of inhabitants/residential area
70-90
72.33747079
0.007553272 0.995191529
90-05
6.740035
7.310378
0.086547
R
2 value 70-90
0.97574884
R
2 value 90-05
0.99980001
5 Influence of Political Regime Change to Land Use Development in Urban. . .
101