Page 41 - Contributed Paper Session (CPS) - Volume 2

CPS1414 Bashiru. I.I S. at el.
variables to vary locally. Prior to the GWLM modelling, a GLM which
assumes fixed parameter was first specified. A Moran I test was then
used to determine the presence of spatial autocorrelation in the
residuals of the model.

2.3 Moran’s I test for spatial autocorrelation

n
n   n w ij x ( i − x)( x j − x)
Moran’s I = = i 1 = j 1 (6)
SumW  n = i 1 x ( i − x) 2

where n is the number of cases indexed by i and j , x is the variable
of interest, x is the mean of x ' , w is the weight between cases i
s
i ij
and j , and SumW is the sum of all w '
s
ij
n
n
SumW =  w (7)
ij
= i 1 = j 1

3. Results and Discussions

Table 2: parameter estimates of the conventional model (GLM)
Parameter Estimate Std. t-value P-value
error
Intercept -0.2069 0.1784 -1.1600 0.2479
Schild 0.0017 0.0009 1.8690 0.0634
*
Aged 0.0228*
0.0026 0.0011 2.2980 *
log(Const) 0.0259*
-0.0230 0.0102 -2.2490 *
log(Agric) 0.0335*
0.0416 0.0194 2.1440 *
HSabove
-0.0002 0.0008 -0.2090 0.8348
Unemployed -0.0015 0.0012 -1.2440 0.2152
Poor 0.0006 0.0005 1.1440 0.2545
log(Bschool) -0.0144 0.0292 -0.4930 0.6227
Mining 0.0000 0.0000 0.3610 0.7188
R 0.25
2
Adj. R 2

0.21
AICc -321.26
Moran’s I
Test 0.001
p≤ 0.001 ***, p≤ 0.05 **, p≤ 0.010*

30 | I S I W S C 2 0 1 9