Sensitivity analysis for choosing minimum number of SSUs per PSU (no sampling frame available)

This function is similar to the function sensitivity_min_SSU, as it allows to analyse the different results in terms of first stage size (number of PSUs) and second stage size (number of SSUs), when varying the values of the minimum number of SSU per single PSU) The name of the parameter has to be given, together with the minimum and maximum value. Differently from sensitivity_min_SSU, it requires in input all the outputs already prepared by prepareInputToAllocation1 or prepareInputToAllocation2.

sensitivity_min_SSU2 (strata, 
                      des_file,
                      psu_file,
                      rho,
                      effst,
                      errors, 
                      min, 
                      max, 
                      plot)

Arguments

strata

Data frame of survey strata, for more details see, e.g.,strata.

des_file

Data frame containing information on sampling design variables, for more details see, e.g.,design.

psu_file

Data frame containing information on primary stage units stratification, for more details see, e.g.,PSU_strat.

rho

Data frame of survey strata, for more details see, e.g.,rho.

effst

Data frame of survey strata, for taking into account the estimator effect on each variable, for more details see, e.g.,effst.

errors

Data frame of coefficients of variation for each domain, for more details see, e.g.,errors.

min

starting value for the minimum value of SSUs per PSU.

max

ending value for the minimum value of SSUs per PSU.

plot

plot (TRUE/FALSE) the final result.

Value

A list containing the (i) vector of allocated PSUs in the iterations and (ii) the vector of allocated SSUs in the iterations

Author

Giulio Barcaroli

Examples

if (FALSE) {
library(readr)
samp <- read_rds("https://github.com/barcaroli/R2BEAT_workflows/blob/master/sample.rds?raw=true")
pop <- read_rds("https://github.com/barcaroli/R2BEAT_workflows/blob/master/pop.RDS?raw=true")
library(R2BEAT)
str(samp)
## Sample design description
library(ReGenesees)
samp$stratum_2 <- as.factor(samp$stratum_2)
sample.des <- e.svydesign(samp, 
                          ids= ~ municipality + id_hh, 
                          strata = ~ stratum_2, 
                          weights = ~ weight,
                          self.rep.str = ~ SR,
                          check.data = TRUE)
## Find and collapse lonely strata
ls <- find.lon.strata(sample.des)
if (!is.null(ls)) sample.des <- collapse.strata(sample.des)
## Calibration with known totals
totals <- pop.template(sample.des,
             calmodel = ~ sex : cl_age, 
             partition = ~ region)
totals <- fill.template(pop, totals, mem.frac = 10)
sample.cal <- e.calibrate(sample.des, 
                          totals,
                          calmodel = ~ sex : cl_age, 
                          partition = ~ region,
                          calfun = "logit",
                          bounds = c(0.3, 2.6), 
                          aggregate.stage = 2,
                          force = FALSE)
samp_frame <- pop
RGdes <- sample.des
RGcal <- sample.cal
strata_var <- c("stratum")      
target_vars <- c("income_hh",
                 "active",
                 "inactive",
                 "unemployed")   
weight_var <- "weight"
deff_var <- "stratum"            
id_PSU <- c("municipality")      
id_SSU <- c("id_hh")             
domain_var <- c("region") 
delta <- 1                   
minimum <- 50     
inp <- prepareInputToAllocation2 ( 
  samp_frame,
  RGdes, 
  RGcal, 
  id_PSU, 
  id_SSU, 
  strata_var, 
  target_vars, 
  deff_var, 
  domain_var,
  delta,
  minimum) 
cv <- as.data.frame(list(DOM=c("DOM1","DOM2"),
                         CV1=c(0.02,0.03),
                         CV2=c(0.03,0.06),
                         CV3=c(0.03,0.06),
                         CV4=c(0.05,0.08)))
sens <- sensitivity_min_SSU2 (strata = inp$strata, 
                         des_file = inp$des_file,
                         psu_file = inp$psu_file,
                         rho = inp$rho,
                         effst = inp$effst,
                         errors = cv, 
                         min = 15, 
                         max = 25, 
                         plot = TRUE) 
}