varpoord: Estimation of the variance and deff for sample surveys for indicators on social exclusion and poverty

Description

Computes the estimation of the variance for indicators on social exclusion and poverty.

Usage

varpoord(Y, w_final, age=NULL, pl085=NULL, month_at_work=NULL,
                     Y_den=NULL, Y_thres = NULL, wght_thres = NULL,
                 ID_household, id = NULL, H, PSU, N_h, fh_zero = FALSE,
                 PSU_level=TRUE, sort = NULL, Dom = NULL, period = NULL,
                 gender = NULL, dataset = NULL, X = NULL, periodX = NULL,
                 X_ID_household = NULL, ind_gr = NULL, g = NULL, datasetX = NULL,
                 q, percentage = 60, order_quant = 50, alpha = 20,
                 confidence = 0.95, outp_lin = FALSE, outp_res = FALSE,
                 several.ok = FALSE, type = "linrmpg")

Arguments

Study variable (for example equalized disposable income or gross pension income). One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value

w_final

Weight variable. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as the column count of

age

Age variable. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as the column count of

pl085

Retirement variable (Number of months spent in retirement or early retirement). One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (

month_at_work

Variable for total number of month at work (sum of the number of months spent at full-time work as employee, number of months spent at part-time work as employee, number of months spent at full-time work as self-employed (including family worker), number

Y_den

Denominator variable (for example gross individual earnings). One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vect

Y_thres

Variable (for example equalized disposable income) used for computation and linearization of poverty threshold. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with on

wght_thres

Weight variable used for computation and linearization of poverty threshold. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (len

ID_household

Variable for household ID codes. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as the

Optional variable for unit ID codes. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as

The unit stratum variable. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as the column

PSU

Primary sampling unit variable. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as the c

N_h

optional; either a matrix giving the first column - stratum, but the second column - the total of the population in each stratum.

fh_zero

by default FALSE; fh is calculated as division of n_h and N_h in each strata, if true, fh value is zero in each strata.

PSU_level

by default TRUE; if PSU_level is true, in each strata fh is calculated as division of count of PSU in sample (n_h) and count of PSU in frame(N_h). if PSU_level is false, in each strata fh is calculated as division of count of units in sample (n_h) and cou

sort

Optional variable to be used as tie-breaker for sorting. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector ha

Dom

Optional variables used to define population domains. If supplied, variables is calculated for each domain. An object convertible to data.frame or variable names as character vector, column numbers or logical vector (length of the vector has

period

Optional variable for survey period. If supplied, variables is calculated for each time period. Object convertable to data.frame or variable names as character, column numbers or logical vector (length of the vector has to be the same as the

gender

Numerical variable for gender, where 1 is for males, but 2 is for females. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (lengt

dataset

Optional survey data object convertible to data.frame.

Optional matrix of the auxiliary variables for the calibration estimator. Object convertible to data.frame or variable names as character, column numbers or logical vector (length of the vector has to be the same as the column count of

periodX

Optional variable of the survey periods. If supplied, residual estimation of calibration is done independently for each time period. Object convertible to data.frame or variable names as character, column numbers or logical vector (length of

X_ID_household

ind_gr

Optional variable by which divided independently auxiliary variables. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of

Optional variable of the g weights. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the vector has to be the same as t

datasetX

Optional survey data object in household level convertible to data.frame.

Variable of the positive values accounting for heteroscedasticity. One dimensional object convertible to one-column data.frame or variable name as character, column number or logical vector with only one TRUE value (length of the

percentage

A numeric value in range $[0,100]$ for $p$ in the formula for poverty threshold computation: $$\frac{p}{100} \cdot Z_{\frac{\alpha}{100}}.$$ For example, to compute poverty threshold equal to 60% of some income quantile, $p$ should be set equal to 60.

order_quant

A numeric value in range $[0,100]$ for $\alpha$ in the formula for poverty threshold computation: $$\frac{p}{100} \cdot Z_{\frac{\alpha}{100}}.$$ For example, to compute poverty threshold equal to some percentage of median income, $\alpha$ should be set

alpha

a numeric value in range $[0,100]$ for the order of the income quantile share ratio (in percentage).

confidence

Optional positive value for confidence interval. This variable by default is 0.95.

outp_lin

Logical value. If TRUE linearized values of the ratio estimator will be printed out.

outp_res

Logical value. If TRUE estimated residuals of calibration will be printed out.

several.ok

Logical value. If type should be allowed to have more than one.

type

a character vector (of length one unless several.ok is TRUE), example "linarpr","linarpt", "lingpg", "linpoormed", "linrmpg", "lingini", "lingini2", "linqsr", "linarr", "linrmi", "all_choices".

Value

A list with objects are returned by the function:
lin_outA data.table containing the linearized values of the ratio estimator with id and PSU.
res_outA data.table containing the estimated residuals of calibration with id and PSU.
all_resultA data.table, which containing variables: respondent_count - the count of respondents, pop_size - the estimated size of population, n_nonzero - the count of respondents, who answers are larger than zero, value - the estimated value, var - the estimated variance, se - the estimated standard error, rse - the estimated relative standard error (coefficient of variation), cv - the estimated relative standard error (coefficient of variation) in percentage, absolute_margin_of_error - the estimated absolute margin of error, relative_margin_of_error - the estimated relative margin of error, CI_lower - the estimated confidence interval lower bound, CI_upper - the estimated confidence interval upper bound, var_srs_HT - the estimated variance of the HT estimator under SRS, var_cur_HT - the estimated variance of the HT estimator under current design, var_srs_ca - the estimated variance of the calibrated estimator under SRS, deff_sam - the estimated design effect of sample design, deff_est - the estimated design effect of estimator, deff - the overall estimated design effect of sample design and estimator, n_eff - the effective sample size.

References

Eric Graf and Yves Tille, Variance Estimation Using Linearization for Poverty and Social Exclusion Indicators, Survey Methodology, June 2014 61 Vol. 40, No. 1, pp. 61-79, Statistics Canada, Catalogue no. 12-001-X, URL http://www.statcan.gc.ca/pub/12-001-x/12-001-x2014001-eng.pdf Guillaume Osier and Emilio Di Meglio. The linearisation approach implemented by Eurostat for the first wave of EU-SILC: what could be done from the second wave onwards? 2012 Guillaume Osier (2009). Variance estimation for complex indicators of poverty and inequality. Journal of the European Survey Research Association, Vol.3, No.3, pp. 167-195, ISSN 1864-3361, URL http://ojs.ub.uni-konstanz.de/srm/article/view/369. Eurostat Methodologies and Working papers, Standard error estimation for the EU-SILC indicators of poverty and social exclusion, 2013, URL http://ec.europa.eu/eurostat/documents/3859598/5927001/KS-RA-13-029-EN.PDF. Jean-Claude Deville (1999). Variance estimation for complex statistics and estimators: linearization and residual techniques. Survey Methodology, 25, 193-203, URL http://www5.statcan.gc.ca/bsolc/olc-cel/olc-cel?lang=eng&catno=12-001-X19990024882. MATTI LANGEL - YVES TILLE, Corrado Gini, a pioneer in balanced sampling and inequality theory. METRON - International Journal of Statistics, 2011, vol. LXIX, n. 1, pp. 45-65, URL ftp://metron.sta.uniroma1.it/RePEc/articoli/2011-1-3.pdf. Morris H. Hansen, William N. Hurwitz, William G. Madow, (1953), Sample survey methods and theory Volume I Methods and applications, 257-258, Wiley. Yves G. Berger, Tim Goedeme, Guillame Osier (2013). Handbook on standard error estimation and other related sampling issues in EU-SILC, URL http://www.cros-portal.eu/content/handbook-standard-error-estimation-and-other-related-sampling-issues-ver-29072013 Working group on Statistics on Income and Living Conditions (2004) Common cross-sectional EU indicators based on EU-SILC; the gender pay gap. EU-SILC 131-rev/04, Eurostat.

Examples

Run this code

data(eusilc)
dataset <- data.frame(1:nrow(eusilc),eusilc)
colnames(dataset)[1] <- "IDd"
dataset1 <- dataset[1:1000,]

aa<-varpoord(Y = "eqIncome", w_final = "rb050",
             Y_thres = NULL, wght_thres = NULL,
             ID_household = "db030", id = "IDd", 
             H = "db040", PSU = "rb030", N_h = NULL,
             sort = NULL, Dom = NULL,
             gender = NULL, X = NULL,
             X_ID_household = NULL, g = NULL,
             datasetX = NULL,
             q = rep(1, if (is.null(datasetX)) 
                        nrow(as.data.frame(H)) else nrow(datasetX)),
             dataset =  dataset1, percentage=60, order_quant=50,
             alpha = 20, confidence = .95, outp_lin = FALSE,
             outp_res = FALSE, several.ok=FALSE, type="linarpt")
aa


aa1 <-varpoord(Y = "eqIncome", w_final = "rb050",
             Y_thres = NULL, wght_thres = NULL,
             ID_household = "db030", id = "IDd", 
             H = "db040", PSU = "rb030", N_h = NULL,
             sort = NULL, Dom = "db040",
             gender = NULL, X = NULL,
             X_ID_household = NULL, g = NULL,
             datasetX = NULL,
             q = rep(1, if (is.null(datasetX)) 
                        nrow(as.data.frame(H)) else nrow(datasetX)),
             dataset =  dataset1, percentage=60, order_quant=50,
             alpha = 20, confidence = .95, outp_lin = FALSE,
             outp_res = FALSE, several.ok=FALSE, type="linarpt")
aa1

aa <- varpoord(Y = "eqIncome", w_final = "rb050",
             Y_thres = NULL, wght_thres = NULL,
             ID_household = "db030", id = "IDd", 
             H = "db040", PSU = "rb030", N_h = NULL,
             sort = NULL, Dom = "db040",
             gender = NULL, X = NULL,
             X_ID_household = NULL, g = NULL,
             datasetX = NULL,
             q = rep(1, if (is.null(datasetX)) 
                        nrow(as.data.frame(H)) else nrow(datasetX)),
             dataset =  dataset, percentage=60, order_quant=50,
             alpha = 20, confidence = .95, outp_lin = TRUE,
             outp_res = TRUE, several.ok=FALSE, type="all_choices")
aa$lin_out[20:40]
aa$res_out[20:40]

Run the code above in your browser using DataLab