Table 3

Multivariate predictors of annual Harris County multifamily submersions Poisson-Lognormal-CAR model coefficients and 95% credible intervals (N=786 census tracts)

DepVar: log of annual number of multifamily submersions per 100 per tract
Model 1Model 2
N786786
Df775779
Number of samples200 000200 000
Number of ‘burn in’ samples100 000100 000
Log likelihood−6022.7−6008.5
AIC12 067.412 031.1
BIC12 118.812 063.7
Deviance11 023.0****11 007.8****
Mean absolute deviation2.622.33
Mean-squared error201.174.6
Dispersion multiplier44.8****40.6****
Coefficient(95% credible interval)Coefficient(95% credible interval)
Modeled parameters
Constant−15.5845 (−22.2055 to −9.9184)−13.9681(−17.8787 to −10.7023)
Socioeconomic variables
African-American population0.0008(−0.0002 to 0.0018)
Asian population0.0031(0.0009 to 0.0055)
Hispanic population0.0005(−0.0003 to 0.0013)
Median household income ($'000)−0.00005(−0.0001 to 0.00 004)
Exposure variables
Number of multifamily pools0.7669(0.4301 to 1.1507)0.6825(0.3748 to 1.0107)
Number of children, ages 0–17, in multifamily residences0.0023(−0.00001 to 0.0048)0.0038(0.0019 to 0.0059)
Statistical controls
Area of tract (m2)−0.0047(−0.3178 to 0.2876)−0.0355(−0.4086 to 0.2171)
Distance to city centre (m)0.0504(−0.1704 to 0.2444)0.0428(−0.1138 to 0.2046)
Unmeasured spatial effect
Spatial autocorrelation (Φ)−0.0015(−0.0711 to 0.0551)−0.0031(−0.0760 to 0.0467)
  • ****p≤0.0001.

  • [Bold] represents significant beyond the 0.0001 credible interval. AIC, Aikaike information criterion; BIC, Bayesian information criterion; CAR, Conditional Autoregressive; Df, degrees of freedom.