Survey-weighted zero-inflated Poisson model

svyglm.zip() is an extension to the survey-package to fit survey-weighted zero-inflated Poisson models. It uses svymle to fit sampling-weighted maximum likelihood estimates, based on starting values provided by zeroinfl.

svyglm.zip(formula, design, ...)

Arguments

formula: An object of class formula, i.e. a symbolic description of the model to be fitted. See 'Details' in zeroinfl.
design: An object of class svydesign, providing a specification of the survey design.
...: Other arguments passed down to zeroinfl.

Value

An object of class svymle and svyglm.zip, with some additional information about the model.

Details

Code modified from https://notstatschat.rbind.io/2015/05/26/zero-inflated-poisson-from-complex-samples/.

Examples

if (require("survey")) {
  data(nhanes_sample)
  set.seed(123)
  nhanes_sample$malepartners <- rpois(nrow(nhanes_sample), 2)
  nhanes_sample$malepartners[sample(1:2992, 400)] <- 0

  # create survey design
  des <- svydesign(
    id = ~SDMVPSU,
    strat = ~SDMVSTRA,
    weights = ~WTINT2YR,
    nest = TRUE,
    data = nhanes_sample
  )

  # fit negative binomial regression
  fit <- svyglm.zip(
    malepartners ~ age + factor(RIDRETH1) | age + factor(RIDRETH1),
    des
  )

  # print coefficients and standard errors
  fit
}
#>                    term estimate std.error conf.low conf.high    p.value
#> 2                   age   0.0149    0.0354   0.9469    1.0879 0.6745    
#> 3     factor(RIDRETH1)2   0.0185    0.0754   0.8787    1.1810 0.8062    
#> 4     factor(RIDRETH1)3  -0.0449    0.0284   0.9043    1.0107 0.1133    
#> 5     factor(RIDRETH1)4  -0.0240    0.0276   0.9250    1.0305 0.3843    
#> 6     factor(RIDRETH1)5   0.0371    0.0617   0.9197    1.1712 0.5470    
#> 7        tp.(Intercept)  -1.6694    0.4717   0.0747    0.4748 <0.001 ***
#> 8                tp.age  -0.0333    0.2831   0.5553    1.6848 0.9064    
#> 9  tp.factor(RIDRETH1)2   0.1548    0.2571   0.7053    1.9323 0.5472    
#> 10 tp.factor(RIDRETH1)3  -0.3969    0.2111   0.4446    1.0169 0.0601 .  
#> 11 tp.factor(RIDRETH1)4  -0.2330    0.3050   0.4357    1.4402 0.4450    
#> 12 tp.factor(RIDRETH1)5  -0.3303    0.4744   0.2836    1.8214 0.4863    
#> 
#> Showing robust standard errors on link-scale (untransformed).