| Type: | Package |
| Title: | Semiparametric Model-Assisted Estimation in Finite Populations |
| Version: | 0.1.3 |
| Maintainer: | Carlos Alberto Cardozo Delgado <cardozorpackages@gmail.com> |
| Description: | It is a framework to fit semiparametric regression estimators for the total parameter of a finite population when the interest variable is asymmetric distributed. The main references for this package are Sarndal C.E., Swensson B., and Wretman J. (2003,ISBN: 978-0-387-40620-6, "Model Assisted Survey Sampling." Springer-Verlag) Cardozo C.A, Paula G.A. and Vanegas L.H. (2022) "Generalized log-gamma additive partial linear mdoels with P-spline smoothing", Statistical Papers. Cardozo C.A and Alonso-Malaver C.E. (2022). "Semi-parametric model assisted estimation in finite populations." In preparation. |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.2.3 |
| Suggests: | survey |
| Imports: | gamlss, gamlss.dist, TeachingSampling, methods, dplyr, caret, magrittr |
| NeedsCompilation: | no |
| Packaged: | 2023-04-10 21:59:20 UTC; CARLOS |
| Author: | Carlos Alberto Cardozo Delgado [aut, cre, cph], Carlos E. Alonso-Malaver [aut] |
| Repository: | CRAN |
| Date/Publication: | 2023-04-11 05:40:02 UTC |
Semiparametric Model-Assisted Estimation under a Bernoulli Sampling Design
Description
sreg_ber is used to estimate the total parameter of a finite population generated from a semi-parametric generalized gamma population under a Bernoulli sampling design.
Usage
sreg_ber(location_formula, scale_formula, data, pi, ...)
Arguments
location_formula |
a symbolic description of the systematic component of the location model to be fitted. |
scale_formula |
a symbolic description of the systematic component of the scale model to be fitted. |
data |
a data frame, list containing the variables in the model. |
pi |
numeric, represents the first order probability. Default value is 0.5. |
... |
further parameters accepted by caret and survey functions. |
Value
sampling_design is the name of the sampling design used in the estimation process.
N is the population size.
n is the random sample size used in the estimation process.
first_order_probabilities vector of the first order probabilities used in the estimation process.
sample is the random sample used in the estimation process.
estimated_total_y_sreg is the SREG estimate of the total parameter of the finite population.
Author(s)
Carlos Alberto Cardozo Delgado <cardozorpackages@gmail.com>
References
Cardozo C.A, Alonso C. (2021) Semi-parametric model assisted estimation in finite populations. In preparation.
Cardozo C.A., Paula G., and Vanegas L. (2022). Generalized log-gamma additive partial linear models with P-spline smoothing. Statistical Papers.
Sarndal C.E., Swensson B., and Wretman J. (2003). Model Assisted Survey Sampling. Springer-Verlag.
Examples
#This example use the data set 'apipop' of the survey package.
library(sregsurvey)
library(survey)
library(magrittr)
library(dplyr)
library(gamlss)
data(api)
attach(apipop)
Apipop <- filter(apipop,full!= 'NA')
Apipop <- filter(Apipop, stype == 'H')
Apipop <- Apipop %>% dplyr::select(api00,grad.sch,full)
fit <- sreg_ber(api00 ~ pb(grad.sch), scale_formula = ~ full - 1, data= Apipop, pi=0.2)
fit
# The total population value is
true_total <- sum(Apipop$api00)
# The estimated relative bias in percentage is
round(abs((fit$estimated_total_y_sreg - true_total)/true_total),3)*100
Semiparametric Model-Assisted Estimation under a Proportional to Size Sampling Design
Description
sreg_pips is used to estimate the total parameter of a finite population generated from a semi-parametric generalized gamma population under a proportional to size without-replacement sampling design.
Usage
sreg_pips(location_formula, scale_formula, data, x, n, ...)
Arguments
location_formula |
a symbolic description of the systematic component of the location model to be fitted. |
scale_formula |
a symbolic description of the systematic component of the scale model to be fitted. |
data |
a data frame, list containing the variables in the model. |
x |
vector, an auxiliary variable to calculate the inclusion probabilities of each unit. |
n |
numeric, sample size. |
... |
further parameters accepted by caret and survey functions. |
Value
sampling_design is the name of the sampling design used in the estimation process.
N is the population size.
n is the sample size used in the estimation process.
first_order_probabilities vector of the first order probabilities used in the estimation process.
sample is the random sample used in the estimation process.
estimated_total_y_sreg is the SREG estimate of the total parameter of the finite population.
Author(s)
Carlos Alberto Cardozo Delgado <cardozorpackages@gmail.com>
References
Cardozo C.A, Alonso C. (2021) Semi-parametric model assisted estimation in finite populations. In preparation.
Cardozo C.A., Paula G., and Vanegas L. (2022). Generalized log-gamma additive partial linear models with P-spline smoothing. Statistical Papers.
Sarndal C.E., Swensson B., and Wretman J. (2003). Model Assisted Survey Sampling. Springer-Verlag.
Examples
library(sregsurvey)
library(survey)
library(dplyr)
library(gamlss)
data(api)
attach(apipop)
Apipop <- filter(apipop,full!= 'NA')
Apipop <- filter(Apipop, stype == 'H')
Apipop <- Apipop %>% dplyr::select(api00,grad.sch,full,api99)
n=ceiling(0.2*dim(Apipop)[1])
aux_var <- Apipop %>% dplyr::select(api99)
fit <- sreg_pips(api00 ~ pb(grad.sch), scale_formula = ~ full - 1, data= Apipop, x= aux_var, n=n)
fit
# The total population value is
true_total <- sum(Apipop$api00)
# The estimated relative bias in percentage is
round(abs((fit$estimated_total_y_sreg - true_total)/true_total),3)*100
Semiparametric Model-Assisted Estimation under a Poisson Sampling Design
Description
sreg_poisson is used to estimate the total parameter of a finite population generated from a semi-parametric generalized gamma population under a Poisson sampling design.
Usage
sreg_poisson(location_formula, scale_formula, data, pis, ...)
Arguments
location_formula |
a symbolic description of the systematic component of the location model to be fitted. |
scale_formula |
a symbolic description of the systematic component of the scale model to be fitted. |
data |
a data frame, list containing the variables in the model. |
pis |
numeric vector, first order inclusion probabilities. Default value 0.1 for each element. |
... |
further parameters accepted by caret and survey functions. |
Value
sampling_design is the name of the sampling design used in the estimation process.
N is the population size.
n is the random sample size used in the estimation process.
first_order_probabilities vector of the first order probabilities used in the estimation process.
sample is the random sample used in the estimation process.
estimated_total_y_sreg is the SREG estimate of the total parameter of the finite population.
Author(s)
Carlos Alberto Cardozo Delgado <cardozorpackages@gmail.com>
References
Cardozo C.A, Alonso C. (2021) Semi-parametric model assisted estimation in finite populations. In preparation.
Cardozo C.A., Paula G., and Vanegas L. (2022). Generalized log-gamma additive partial linear models with P-spline smoothing. Statistical Papers.
Sarndal C.E., Swensson B., and Wretman J. (2003). Model Assisted Survey Sampling. Springer-Verlag.
Examples
library(sregsurvey)
library(survey)
library(dplyr)
library(gamlss)
data(api)
attach(apipop)
Apipop <- filter(apipop,full!= 'NA')
Apipop <- filter(Apipop, stype == 'H')
Apipop <- Apipop %>% dplyr::select(api00,grad.sch,full)
fit <- sreg_poisson(api00 ~ pb(grad.sch), scale_formula = ~ full - 1, data= Apipop)
fit
# The total population value is
true_total <- sum(Apipop$api00)
# The estimated relative bias in percentage is
round(abs((fit$estimated_total_y_sreg - true_total)/true_total),3)*100
Semiparametric Model-Assisted Estimation under a Simple Random Sampling Without Replace Sampling Design
Description
sreg_srswr is used to estimate the total parameter of a finite population generated from a semi-parametric generalized gamma population under a simple random sampling without-replacement sampling design.
Usage
sreg_srswr(
location_formula,
scale_formula,
data,
fraction,
format = "COMPLETE",
...
)
Arguments
location_formula |
a symbolic description of the systematic component of the location model to be fitted. |
scale_formula |
a symbolic description of the systematic component of the scale model to be fitted. |
data |
a data frame, list containing the variables in the model. |
fraction |
numeric, represents a fraction of the size of the population. Default value is 0.2. |
format |
character, represents the type of summary of the methodology, 'SIMPLE' or 'COMPLETE'. Default value is 'COMPLETE'. |
... |
further parameters accepted by caret and survey functions. |
Value
sampling_design is the name of the sampling design used in the estimation process.
N is the population size.
n is the fixed sample size used in the estimation process.
first_order_probabilities vector of the first order probabilities used in the estimation process.
sample is the random sample used in the estimation process.
estimated_total_y_sreg is the SREG estimate of the total parameter of the finite population.
Author(s)
Carlos Alberto Cardozo Delgado <cardozorpackages@gmail.com>
References
Cardozo C.A, Alonso C. (2021) Semi-parametric model assisted estimation in finite populations. In preparation.
Cardozo C.A., Paula G., and Vanegas L. (2022). Generalized log-gamma additive partial linear models with P-spline smoothing. Statistical Papers.
Sarndal C.E., Swensson B., and Wretman J. (2003). Model Assisted Survey Sampling. Springer-Verlag.
Examples
library(sregsurvey)
library(survey)
library(dplyr)
library(gamlss)
data(api)
attach(apipop)
Apipop <- filter(apipop,full!= 'NA')
Apipop <- filter(Apipop, stype == 'H')
Apipop <- Apipop %>% dplyr::select(api00,grad.sch,full)
fit <- sreg_srswr(api00 ~ pb(grad.sch), scale_formula = ~ full - 1, data= Apipop, fraction=0.25)
# The total population value is
true_total <- sum(Apipop$api00)
# The estimated relative bias in percentage is
round(abs((fit$estimated_total_y_sreg - true_total)/true_total),3)*100
Semiparametric Model-Assisted Estimation under a Stratified Sampling with Simple Random Sampling Without Replace in each stratum.
Description
sreg_stsi is used to estimate the total parameter of a finite population generated from a semi-parametric generalized gamma population
under a stratified sampling with simple random sampling without-replacement in each stratum.
Usage
sreg_stsi(
location_formula,
scale_formula,
stratum,
data,
n,
ss_sizes,
allocation_type = "PA",
aux_x,
...
)
Arguments
location_formula |
a symbolic description of the systematic component of the location model to be fitted. |
scale_formula |
a symbolic description of the systematic component of the scale model to be fitted. |
stratum |
vector, represents the strata of each unit in the population |
data |
a data frame, list containing the variables in the model. |
n |
integer, represents a fixed sample size. |
ss_sizes |
vector, represents a vector with the sample size in each stratum. |
allocation_type |
character, there is two choices, proportional allocation, 'PA', and x-optimal allocation,'XOA'. By default is a 'PA', Sarndal et. al. (2003). |
aux_x |
vector, represents an auxiliary variable to help to calculate the sample sizes by the x-optimum allocation method, Sarndal et. al. (2003). This option is validated only when the argument allocation_type is equal to 'XOA'. |
... |
further parameters accepted by caret and survey functions. |
Value
sampling_design is the name of the sampling design used in the estimation process.
N is the population size.
H is the number of strata.
Ns is the population strata sizes.
allocation_type is the method used to calculate sample strata sizes.
global_n is the global sample size used in the estimation process.
first_order_probabilities vector of the first order probabilities used in the estimation process.
sample is the random sample used in the estimation process.
estimated_total_y_sreg is the SREG estimate of the total parameter of the finite population.
Author(s)
Carlos Alberto Cardozo Delgado <cardozorpackages@gmail.com>
References
Cardozo C.A, Alonso C. (2021) Semi-parametric model assisted estimation in finite populations. In preparation.
Cardozo C.A., Paula G., and Vanegas L. (2022). Generalized log-gamma additive partial linear models with P-spline smoothing. Statistical Papers.
Sarndal C.E., Swensson B., and Wretman J. (2003). Model Assisted Survey Sampling. Springer-Verlag.
Examples
library(sregsurvey)
library(survey)
library(dplyr)
library(magrittr)
library(gamlss)
data(api)
attach(apipop)
Apipop <- filter(apipop,full!= 'NA')
Apipop <- Apipop %>% dplyr::select(api00,grad.sch,full,stype)
dim(Apipop)
fit <- sreg_stsi(api00~ pb(grad.sch), scale_formula =~ full-1, n=400, stratum='stype', data=Apipop)
fit
# The total population value is
true_total <- sum(Apipop$api00)
# The estimated relative bias in percentage is
round(abs((fit$estimated_total_y_sreg - true_total)/true_total),3)*100