This vignette demonstrates the implementation of treed distributed lag non-linear model (TDLNM). More details can be found in Mork and Wilson (2021) <doi: 10.1093/biostatistics/kxaa051>.
Load data
Simulated data is available on GitHub. It can be loaded with the following code.
sbd_dlmtree <- get_sbd_dlmtree()Data preparation
# Response and covariates
sbd_cov <- sbd_dlmtree %>%
select(bwgaz, ChildSex, MomAge, GestAge, MomPriorBMI, Race,
Hispanic, MomEdu, SmkAny, Marital, Income,
EstDateConcept, EstMonthConcept, EstYearConcept)
# Exposure data
sbd_exp <- list(PM25 = sbd_dlmtree %>% select(starts_with("pm25_")),
TEMP = sbd_dlmtree %>% select(starts_with("temp_")),
SO2 = sbd_dlmtree %>% select(starts_with("so2_")),
CO = sbd_dlmtree %>% select(starts_with("co_")),
NO2 = sbd_dlmtree %>% select(starts_with("no2_")))
sbd_exp <- sbd_exp %>% lapply(as.matrix)Fitting the model
tdlnm.fit <- dlmtree(formula = bwgaz ~ ChildSex + MomAge + MomPriorBMI +
Race + Hispanic + SmkAny + EstMonthConcept,
data = sbd_cov,
exposure.data = sbd_exp[["TEMP"]],
dlm.type = "nonlinear",
family = "gaussian",
tdlnm.exposure.splits = 20,
n.burn = 2500, n.iter = 10000, n.thin = 5)
#> Preparing data...
#>
#> Running TDLNM:
#> Burn-in % complete
#> [0--------25--------50--------75--------100]
#> ''''''''''''''''''''''''''''''''''''''''''
#> MCMC iterations (est time: 28 seconds)
#> [0--------25--------50--------75--------100]
#> ''''''''''''''''''''''''''''''''''''''''''
#> Compiling results...Model fit summary
tdlnm.sum <- summary(tdlnm.fit)
#> Centered DLNM at exposure value 0
tdlnm.sum
#> ---
#> TDLNM summary
#>
#> Model run info:
#> - bwgaz ~ ChildSex + MomAge + MomPriorBMI + Race + Hispanic + SmkAny + EstMonthConcept
#> - sample size: 10,000
#> - family: gaussian
#> - 20 trees
#> - 2500 burn-in iterations
#> - 10000 post-burn iterations
#> - 5 thinning factor
#> - 0.95 confidence level
#>
#> Fixed effect coefficients:
#> Mean Lower Upper
#> (Intercept) 0.160 -0.913 1.180
#> *ChildSexM -2.105 -2.127 -2.086
#> MomAge 0.001 -0.001 0.002
#> *MomPriorBMI -0.021 -0.023 -0.019
#> RaceAsianPI 0.026 -0.103 0.152
#> RaceBlack 0.031 -0.094 0.157
#> Racewhite 0.012 -0.112 0.133
#> *HispanicNonHispanic 0.256 0.233 0.278
#> *SmkAnyY -0.396 -0.442 -0.351
#> *EstMonthConcept2 0.117 0.035 0.202
#> *EstMonthConcept3 0.231 0.100 0.356
#> *EstMonthConcept4 0.369 0.201 0.533
#> *EstMonthConcept5 0.498 0.318 0.681
#> *EstMonthConcept6 0.452 0.275 0.639
#> *EstMonthConcept7 0.387 0.214 0.573
#> *EstMonthConcept8 0.238 0.072 0.420
#> *EstMonthConcept9 0.264 0.099 0.437
#> *EstMonthConcept10 0.159 0.014 0.308
#> *EstMonthConcept11 0.127 0.013 0.237
#> EstMonthConcept12 0.020 -0.054 0.094
#> ---
#> * = CI does not contain zero
#>
#> DLNM effect:
#> range = [-0.042, 0.059]
#> signal-to-noise = 0.405
#> critical windows: 4-6,10-34
#>
#> residual standard errors: 0.004Exposure-time surface
plot(tdlnm.sum,
main = "Plot title",
xlab = "Time axis label",
ylab = "Exposure-concentration axis label",
flab = "Effect color label")
