Table 4.

Results for the fixed effects spatial error model (SEA) and the fixed effects spatial autocorrelation model (SAR) for the log of infant mortality rate

Spatial fixed effects
Dependent variable: log(infant mortality rate)	FE SEA model		FE SAR model
Number of public health care institutions per 1,000 people	-0.048	***	-0.012
Number of private health care institutions per 1,000 people	-0.166	***	-0.159	***
% of households with access to water	-0.259	***	-0.142	***
% of households with access to sanitation	-0.222	***	-0.160	***
Average income (2000 R$)	-0.001	***	0.000	***
Gini Index	0.244	***	0.083	***
Female illiteracy rate (15 years old or more)	0.187	***	0.305	***
% of urban population	-0.047	*	-0.221	***
Adolescent fertility rate (10-19 years old)	0.168	***	0.139	*
W * Number of public health care institutions per 1,000 people	-0.213	***
W * Number of private health care institutions per 1,000 people	-0.561	***
W * % of households with access to water	-1.000	***
W * % of households with access to sanitation	-0.569	***
atial error autocorrelation (ρ)	0.786	***
Spatial lag of the log(infant mortality rate) (δ)			0.7789	***
R2	0.8875		0.9554
${\mathbf{\sigma }}_{\mathbf{\epsilon }}^{\mathbf{2}}$	0.0194		0.0188
Loglikelihood	5,264.64		5,447.57
AIC	-10,501.28		-10,875.14
Corr-squared	0.7699		0.8045
N	10,977		10,977

* α = 0.10; ** α = 0.05; *** α = 0.01;