Chapter 3 Model

3.2 ADMINISTRATIVE AND SPATIAL DATA

## Reading layer `tl_2017_us_county' from data source `/research-home/gcarrasco/EMMERG_MAP2/_data/GIS/tl_2017_us_county/tl_2017_us_county.shp' using driver `ESRI Shapefile'
## Simple feature collection with 3233 features and 17 fields
## geometry type:  MULTIPOLYGON
## dimension:      XY
## bbox:           xmin: -179.2311 ymin: -14.60181 xmax: 179.8597 ymax: 71.43979
## epsg (SRID):    4269
## proj4string:    +proj=longlat +datum=NAD83 +no_defs

3.4 Model Estimation

## [1] 0
## [1] 0.2171574
## [1] 0.1994246
## [1] 0.3202601
## [1] 0.3214583
## [1] 0.3214238
## # A tibble: 6 x 8
##   formula                     Rsq   DIC     pD `log score`  waic `waic pD`    id
##   <chr>                     <dbl> <dbl>  <dbl>       <dbl> <dbl>     <dbl> <int>
## 1 "fenta ~ 1"               0     8571.   1.01      0.227  8571.      1.01     1
## 2 "fenta ~ 1 + f(index, mo… 0.217 5175. 612.        0.135  5090.    466.       2
## 3 "fenta ~ 1 + f(index, mo… 0.199 4724. 175.        0.124  4695.    141.       3
## 4 "fenta ~ 1 + f(index, mo… 0.320  Inf  Inf         0.0684 2460.    416.       4
## 5 "fenta ~ 1 + f(index, mo… 0.321  Inf  Inf         0.0681 2432.    420.       5
## 6 "fenta ~ 1 + f(index, mo… 0.321  Inf  Inf         0.0680 2432.    419.       6

3.5 Output

3.5.1 Year Random Effects

## 
## Call:
##    c("inla(formula = formula, family = \"binomial\", data = dat1, verbose 
##    = F, ", " control.compute = list(config = T, dic = T, cpo = T, waic = 
##    T), ", " control.predictor = list(link = 1, compute = TRUE), 
##    control.fixed = list(correlation.matrix = T))" ) 
## Time used:
##     Pre = 45.1, Running = 410, Post = 107, Total = 562 
## Fixed effects:
##                mean    sd 0.025quant 0.5quant 0.975quant    mode kld
## (Intercept) -12.136 0.412    -13.025  -12.108    -11.408 -12.061   0
## 
## Linear combinations (derived):
##             ID    mean    sd 0.025quant 0.5quant 0.975quant   mode kld
## (Intercept)  1 -12.135 0.412    -12.963  -12.123    -11.373 -12.08   0
## 
## Random effects:
##   Name     Model
##     index BYM model
##    Year RW1 model
## 
## Model hyperparameters:
##                                             mean       sd 0.025quant 0.5quant
## Precision for index (iid component)     1983.178 1961.581    142.665 1406.314
## Precision for index (spatial component)    0.019    0.001      0.018    0.019
## Precision for Year                         0.281    0.705      0.022    0.115
##                                         0.975quant    mode
## Precision for index (iid component)        7191.43 394.090
## Precision for index (spatial component)       0.02   0.018
## Precision for Year                            1.56   0.044
## 
## Expected number of effective parameters(stdev): 984.53(4.95)
## Number of equivalent replicates : 19.15 
## 
## Deviance Information Criterion (DIC) ...............: Inf
## Deviance Information Criterion (DIC, saturated) ....: Inf
## Effective number of parameters .....................: Inf
## 
## Watanabe-Akaike information criterion (WAIC) ...: 2460.32
## Effective number of parameters .................: 416.14
## 
## Marginal log-Likelihood:  -1511.13 
## CPO and PIT are computed
## 
## Posterior marginals for the linear predictor and
##  the fitted values are computed