Package 'rifreg' reference manual

Title:	Estimate Recentered Influence Function Regression
Description:	Provides functions to compute recentered influence functions (RIF) of a distributional variable at the mean, quantiles, variance, gini or any custom functional of interest. The package allows to regress the RIF on any number of covariates. Generic print, plot and summary functions are also provided. Reference: Firpo, Sergio, Nicole M. Fortin, and Thomas Lemieux. (2009) <doi:10.3982/ECTA6822>. "Unconditional Quantile Regressions.".
Authors:	David Gallusser [aut], Samuel Meier [aut, cre]
Maintainer:	Samuel Meier <[email protected]>
License:	GPL (>= 3)
Version:	0.1.0
Built:	2024-11-28 06:31:28 UTC
Source:	CRAN

Check weights

Description

Helper function to check a weights vector. Makes sure the weights are positive numeric values (not all zeros) and of the same length as the dependent variable dep_var. Replaces all NAs with 0 and sets all weights to 1 if weights is set to NULL.

Usage

check_weights(dep_var, weights)
check_weights(dep_var, weights)

Arguments

`dep_var`	dependent variable of distributional function. Can be any discrete or continuous vector of length 1 or more.
`weights`	positive numeric vector of `length(dep_var)` containing the weights or `NULL`.

Value

positive numeric vector of length(dep_var) containing the checked weights. If weights = NULL, all weights are set to 1.

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
check_weights(dep_var, weights)

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
check_weights(dep_var, weights)

Generalized Lorenz ordinates

Description

Compute the generalized Lorenz ordinates of dep_var (i.e. the share of total income observations up to each value in dep_var amass scaled by the mean income).

Usage

compute_generalized_lorenz_ordinates(dep_var, weights)
compute_generalized_lorenz_ordinates(dep_var, weights)

Arguments

`dep_var`	dependent variable of a distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`.

Value

thes generalized Lorenz ordinates for a vector dep_var.

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
generalized_lorenz_ordinates <-
  compute_generalized_lorenz_ordinates(
    dep_var = dep_var,
    weights = weights
  )

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
generalized_lorenz_ordinates <-
  compute_generalized_lorenz_ordinates(
    dep_var = dep_var,
    weights = weights
  )

Compute Gini coefficient

Description

Compute a weighted Gini coefficient by integrating the generalized Lorenz curve.

Usage

compute_gini(dep_var, weights)
compute_gini(dep_var, weights)

Arguments

`dep_var`	values of a non-negative continuous variable
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`.

Value

The numeric value indicating the weighted Gini coefficient of the the dependent variable.

References

Firpo, Sergio P., Nicole M. Fortin, and Thomas Lemieux. 2018. “Decomposing Wage Distributions Using Recentered Influence Function Regressions.” Econometrics 6(2), 28.

Examples


set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
compute_gini(dep_var, weights)

set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
compute_gini(dep_var, weights)

Weighted ECDF value

Description

Compute values of the ECDF for a vector dep_var (i.e. the empirical probability for each observation in dep_var that a value in dep_var is smaller or equal).

Usage

compute_weighted_ecdf(dep_var, weights)
compute_weighted_ecdf(dep_var, weights)

Arguments

`dep_var`	dependent variable of a distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`.

Value

the values of ECDF for a vector dep_var.

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
value_of_ecdf <-
  compute_weighted_ecdf(
    dep_var = dep_var,
    weights = weights
  )

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
value_of_ecdf <-
  compute_weighted_ecdf(
    dep_var = dep_var,
    weights = weights
  )

Estimate Recentered Influence Functions

Description

This function estimates the recentered influence function (RIF) of a chosen distributional statistic (e.g. quantiles, variance or gini).

Usage

get_rif(
  dep_var,
  weights = NULL,
  statistic,
  probs = NULL,
  custom_rif_function = NULL,
  ...
)
get_rif(
  dep_var,
  weights = NULL,
  statistic,
  probs = NULL,
  custom_rif_function = NULL,
  ...
)

Arguments

`dep_var`	dependent variable of distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL)` is equivalent to `weights = rep(1, length(dep_var))`.
`statistic`	string containing the distributional statistic for which to compute the RIF. Can be one of "mean", "variance", "quantiles", "gini", "interquantile_range", "interquantile_ratio", or "custom". If "custom" is selected a `custom_rif_function` needs to be provided.
`probs`	a vector of length 1 or more with quantile positions to calculate the RIF. Each quantile is indicated with value between 0 and 1. Only required if `statistic = "quantiles"`.
`custom_rif_function`	the RIF function to compute the RIF of the custom distributional statistic. Default is NULL. Only needs to be provided if `statistic = "custom"`. Every custom_rif_function needs the parameters `dep_var`, `weights` and `probs`. If they are not needed they must be set to NULL in the function definition (e.g. `probs = NULL`). A custom function must return a data frame containing at least a "rif" and "weights" column. See `examples` for further details.
`...`	additional parameters passed to the `custom_rif_function`. Apart from `dep_var`, `weights` and `probs` they must have a different name than the the ones in `rif`. For instance, if you want to pass a parameter `statistic` to the `custom_rif_function`, name it `custom_statistic`.

Value

a data frame with the RIF value for each observation and in the case of several quantiles a column for each quantile.

References

Firpo, Sergio P., Nicole M. Fortin, and Thomas Lemieux. 2009. “Unconditional Quantile Regressions.” Econometrica 77(3): 953–73.

Cowell, Frank A., and Emmanuel Flachaire. 2015. “Statistical Methods for Distributional Analysis.” In Anthony B. Atkinson and François Bourguignon (eds.), Handbook of Income Distribution. Amsterdam: Elsevier.

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
probs <- seq(1:9) / 10
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
rif <- get_rif(
  dep_var = dep_var,
  weights = weights,
  statistic = "quantiles",
  probs = probs
)

# custom function
custom_variance_function <- function(dep_var, weights, probs = NULL) {
  weighted_mean <- weighted.mean(x = dep_var, w = weights)
  rif <- (dep_var - weighted_mean)^2
  rif <- data.frame(rif, weights)
  names(rif) <- c("rif_variance", "weights")
  return(rif)
}

set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)

# custom function top 10% percent income share
# (see Essam-Nassah & Lambert, 2012, and Rios-Avila, 2020)
custom_top_income_share_function <- function(dep_var, weights, probs = 0.1) {
  probs <- 1 - probs
  weighted_mean <- weighted.mean(x = dep_var, w = weights)
  weighted_quantile <- Hmisc::wtd.quantile(x = dep_var, weights = weights, probs = probs)
  lorenz_ordinate <-
    sum(dep_var[which(dep_var <= weighted_quantile)] *
      weights[which(dep_var <= weighted_quantile)]) / sum(dep_var * weights)
  if_lorenz_ordinate <- -(dep_var / weighted_mean) * lorenz_ordinate +
    ifelse(dep_var < weighted_quantile,
      dep_var - (1 - probs) * weighted_quantile,
      probs * weighted_quantile
    ) / weighted_mean
  rif_top_income_share <- (1 - lorenz_ordinate) - if_lorenz_ordinate
  rif <- data.frame(rif_top_income_share, weights)
  names(rif) <- c("rif_top_income_share", "weights")
  return(rif_top_income_share)
}

rif_custom <- get_rif(
  dep_var = dep_var,
  weights = weights,
  statistic = "custom",
  custom_rif_function = custom_variance_function
)

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
probs <- seq(1:9) / 10
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
rif <- get_rif(
  dep_var = dep_var,
  weights = weights,
  statistic = "quantiles",
  probs = probs
)

# custom function
custom_variance_function <- function(dep_var, weights, probs = NULL) {
  weighted_mean <- weighted.mean(x = dep_var, w = weights)
  rif <- (dep_var - weighted_mean)^2
  rif <- data.frame(rif, weights)
  names(rif) <- c("rif_variance", "weights")
  return(rif)
}

set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)

# custom function top 10% percent income share
# (see Essam-Nassah & Lambert, 2012, and Rios-Avila, 2020)
custom_top_income_share_function <- function(dep_var, weights, probs = 0.1) {
  probs <- 1 - probs
  weighted_mean <- weighted.mean(x = dep_var, w = weights)
  weighted_quantile <- Hmisc::wtd.quantile(x = dep_var, weights = weights, probs = probs)
  lorenz_ordinate <-
    sum(dep_var[which(dep_var <= weighted_quantile)] *
      weights[which(dep_var <= weighted_quantile)]) / sum(dep_var * weights)
  if_lorenz_ordinate <- -(dep_var / weighted_mean) * lorenz_ordinate +
    ifelse(dep_var < weighted_quantile,
      dep_var - (1 - probs) * weighted_quantile,
      probs * weighted_quantile
    ) / weighted_mean
  rif_top_income_share <- (1 - lorenz_ordinate) - if_lorenz_ordinate
  rif <- data.frame(rif_top_income_share, weights)
  names(rif) <- c("rif_top_income_share", "weights")
  return(rif_top_income_share)
}

rif_custom <- get_rif(
  dep_var = dep_var,
  weights = weights,
  statistic = "custom",
  custom_rif_function = custom_variance_function
)

Estimate RIF of Gini coefficient

Description

Compute the recentered influence function (RIF) of a weighted Gini coefficient.

Usage

get_rif_gini(dep_var, weights)
get_rif_gini(dep_var, weights)

Arguments

`dep_var`	values of a non-negative continuous dependent variable
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`.

Value

A data frame with one column containing the RIF of the Gini coefficient for each observation.

References

Cowell, Frank A., and Emmanuel Flachaire. 2007. "Income distribution and inequality measurement: The problem of extreme values." Journal of Econometrics, 141(2), 1044-1072.

Firpo, Sergio P., Nicole M. Fortin, and Thomas Lemieux. 2018. “Decomposing Wage Distributions Using Recentered Influence Function Regressions.” Econometrics 6(2), 28.

Monti, Anna Clara. 1991. "The study of the Gini concentration ratio by means of the influence function." Statistica 51(4), 561–577.

Examples


set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
rif_gini <- get_rif_gini(dep_var = dep_var, weights = weights)
rif_gini

gini <- compute_gini(dep_var = dep_var, weights = weights)
all.equal(gini, mean(rif_gini$rif_gini))

set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
rif_gini <- get_rif_gini(dep_var = dep_var, weights = weights)
rif_gini

gini <- compute_gini(dep_var = dep_var, weights = weights)
all.equal(gini, mean(rif_gini$rif_gini))

Estimate RIF of interquantile range

Description

Compute the recentered influence function (RIF) of a weighted interquantile range.

Usage

get_rif_interquantile_range(dep_var, weights, probs, ...)
get_rif_interquantile_range(dep_var, weights, probs, ...)

Arguments

`dep_var`	dependent variable of distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL)` is equivalent to `weights = rep(1, length(dep_var))`.
`probs`	a vector of length 2 with probabilities corresponding to the limits of the interquantile range of interest. The interquantile range is defined as difference between the quantile with the larger probability and the one with the lower probability.
`...`	further arguments passed on to density.

Value

A data frame with one column containing the RIF of the interquantile range for each observation and one column containing the weights.

References

Firpo, Sergio P., Nicole M. Fortin, and Thomas Lemieux. 2018. “Decomposing Wage Distributions Using Recentered Influence Function Regressions.” Econometrics 6(2), 28.

Examples


set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
get_rif_interquantile_range(dep_var, probs = c(0.1, 0.9), weights = weights)

set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
get_rif_interquantile_range(dep_var, probs = c(0.1, 0.9), weights = weights)

Estimate RIF of interquantile ratio

Description

Compute the recentered influence function (RIF) of a weighted interquantile ratio.

Usage

get_rif_interquantile_ratio(dep_var, weights, probs, ...)
get_rif_interquantile_ratio(dep_var, weights, probs, ...)

Arguments

`dep_var`	dependent variable of a distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL)` is equivalent to `weights = rep(1, length(dep_var))`.
`probs`	a vector of length 2 with probabilities corresponding to the quantiles in the ratio's numerator and the denominator. The function defines the interquantile ratio as the ratio between the quantile with the larger probability (numerator) and the quantile with the lower probability (denominator).
`...`	further arguments passed on to density.

Value

A data frame with one column containing the RIF of the interquantile ratio for each observation.

References

Chung, Choe, and Philippe Van Kerm. 2018. "Foreign workers and the wage distribution: What does the infuence function reveal?", Econometrics 6(3), 41.

Examples


set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
get_rif_interquantile_ratio(dep_var, probs = c(0.1, 0.9), weights = weights)

set.seed(123)
dep_var <- rlnorm(100)
weights <- rep(1, 100)
get_rif_interquantile_ratio(dep_var, probs = c(0.1, 0.9), weights = weights)

Estimate RIF at the Mean

Description

Function to estimate the recentered influence function (RIF) at the mean of a weighted distribution of a dependent variable.

Usage

get_rif_mean(dep_var)
get_rif_mean(dep_var)

Arguments

dep_var

dependent variable of a distributional function. Discrete or continuous numeric vector.

Value

A data frame with one column of length(dep_var) containing the RIF at the mean.

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
get_rif_mean(dep_var)

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
get_rif_mean(dep_var)

Estimate RIF at Quantiles

Description

Function to estimate the recentered influence function (RIF) at one or several specified quantiles of a weighted distribution of a dependent variable.

Usage

get_rif_quantiles(dep_var, weights, probs, ...)

get_rif_quantile(dep_var, weights, probs, ...)
get_rif_quantiles(dep_var, weights, probs, ...)

get_rif_quantile(dep_var, weights, probs, ...)

Arguments

`dep_var`	dependent variable of a distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL)` is equivalent to `weights = rep(1, length(dep_var))`.
`probs`	the specific quantile at which to estimate the RIF.
`...`	further arguments passed on to density.

Value

A data frame with the number of columns equaling the length of vector probs and an additional column containing the weights. Each column contains the RIF values at the quantile's probabilities.

Functions

get_rif_quantile(): Helper function to estimate the RIF values at a specific quantile.

Examples

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
probs <- seq(1:9) / 10
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
get_rif_quantiles(dep_var, probs, weights = weights)

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
probs <- seq(1:9) / 10
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
get_rif_quantiles(dep_var, probs, weights = weights)

Estimate RIF of variance

Description

Function to estimate the recentered influence function (RIF) of the variance of a weighted distribution of a dependent variable.

Usage

get_rif_variance(dep_var, weights)
get_rif_variance(dep_var, weights)

Arguments

`dep_var`	dependent variable of a distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL)` is equivalent to `weights = rep(1, length(dep_var))`.

Value

A data frame with one column containing the RIF of the variance for each observation and one column containing the weights.

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
get_rif_variance(dep_var, weights = weights)

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
get_rif_variance(dep_var, weights = weights)

Integrate generalized Lorenz curve

Description

Computes the area under the lorenz curve.

Usage

integrate_generalized_lorenz_curve(dep_var, weights)
integrate_generalized_lorenz_curve(dep_var, weights)

Arguments

`dep_var`	dependent variable of a distributional function. Discrete or continuous numeric vector.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL)` is equivalent to `weights = rep(1, length(dep_var))`.

Value

the size of the area under the lorenz curve (the integrated lorenz curve).

Examples


dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
integrated_lorenz_curve <-
  integrate_generalized_lorenz_curve(
    dep_var = dep_var,
    weights = weights
  )

dep_var <- c(1, 3, 9, 16, 3, 7, 4, 9)
weights <- c(2, 1, 3, 4, 4, 1, 6, 3)
integrated_lorenz_curve <-
  integrate_generalized_lorenz_curve(
    dep_var = dep_var,
    weights = weights
  )

Sample of male wage data from the CPS 1983-1985

Description

A sample of the the Merged Outgoing Rotation Group of the Current Population Survey of 1983, 1984 and 1985 used by Firpo, Fortin & Lemieux (2009). The data contains a selection of 10 variables and a sample of 26,695 observations of male workers – corresponding to a tenth of the original 266,956 observations. See Lemieux (2006) for details on data selection and recoding.

Usage

men8385
men8385

Format

A data frame with 26,695 rows and 10 variables.

wage: Hourly wage in US dollars at constant prices
union: Union status indicator
nonwhite: Non-white indicator
married: Married indicator
education: Factor variable with 6 education levels: high-school graduates (reference), elementary, high-school dropouts , some college, college graduates, post college graduates
experience: Factor variable with 9 potential experience levels, each of five years gap, 20 to 24 years as reference level)
weights: CPS sample weights
age: Age in years
education_in_years: Education in years
experience_in_years: Experience in years

Source

Sergio Firpo, Nicole M. Fortin, and Thomas Lemieux, "Unconditional Quantile Regressions", Econometrica, Vol. 77, No. 3 (May, 2009), pp. 953-973.

Replication files: <https://www.econometricsociety.org/publications/econometrica/2009/05/01/unconditional-quantile-regressions>

Thoms Lemieux, "Increasing Residual Wage Inequality: Composition Effects, Noisy Data, or Rising Demand for Skill?", American Economic Review, Vol. 96, No. 3 (June, 2006), pp. 461-498.

Plot the coefficients of a `rifreg` object

Description

Coefficients are plotted for each quantile and each covariate. Specific covariates can be selected and standard errors displayed if desired.

Usage

## S3 method for class 'rifreg'
plot(
  x,
  varselect = NULL,
  confidence_level = 0.05,
  vcov = sandwich::sandwich,
  ...
)
## S3 method for class 'rifreg'
plot(
  x,
  varselect = NULL,
  confidence_level = 0.05,
  vcov = sandwich::sandwich,
  ...
)

Arguments

`x`	an object of class "rifreg", usually, a result of a call to rifreg with `statistic = "quantiles"`.
`varselect`	vector of length 1 or more containig the names of the covariates to display.
`confidence_level`	numeric value between 0 and 1 (default = 0.95) that defines the confidence interval plotted as a ribbon and defined as `qnorm(confidence_level/2)` * standard error.
`vcov`	Function to estimate covariance matrix of rifreg coefficients if covariance matrix has not been bootstrapped. Per default, heteroscedasticity-consistent (HC) standard errors are calculated using sandwich. Note: These standard errors do not take the variance introduced by estimating RIF into account.
`...`	other parameters to be passed to plotting function. See ggplot for further information.

Value

a "ggplot" containing the coefficients for each (selected) covariate

Examples


rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  probs = seq(0.1, 0.9, 0.1),
  weights = weights
)

plot(rifreg)

plot(rifreg, varselect = c("age", "unionyes"), confidence_level = 0.1)

rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  probs = seq(0.1, 0.9, 0.1),
  weights = weights
)

plot(rifreg)

plot(rifreg, varselect = c("age", "unionyes"), confidence_level = 0.1)

Print method for class "rifreg"

Description

Print method for class "rifreg"

Usage

## S3 method for class 'rifreg'
print(x, ...)
## S3 method for class 'rifreg'
print(x, ...)

Arguments

`x`	an object of class "rifreg", usually, a result of a call to rifreg.
`...`	other parameters to be passed to printing function.

Value

the function print.rifreg() returns the the covariates' coefficients of the RIF regressions derived from the fitted linear model given in object x.

Examples


rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  probs = seq(0.1, 0.9, 0.1),
  weights = weights
)

print(rifreg)

rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  probs = seq(0.1, 0.9, 0.1),
  weights = weights
)

print(rifreg)

RIF regression

Description

Estimate a recentered influence function (RIF) regression for a distributional statistic of interest.

Usage

rifreg(
  formula,
  data,
  statistic = "quantiles",
  weights = NULL,
  probs = c(1:9)/10,
  custom_rif_function = NULL,
  na.action = na.omit,
  bootstrap = FALSE,
  bootstrap_iterations = 100,
  cores = 1,
  ...
)
rifreg(
  formula,
  data,
  statistic = "quantiles",
  weights = NULL,
  probs = c(1:9)/10,
  custom_rif_function = NULL,
  na.action = na.omit,
  bootstrap = FALSE,
  bootstrap_iterations = 100,
  cores = 1,
  ...
)

Arguments

`formula`	an object of class "formula". See lm for further details.
`data`	a data frame containing the variables in the model.
`statistic`	string containing the distributional statistic for which to compute the RIF. Can be one of "quantiles", "mean", "variance", "gini", "interquantile_range", "interquantile_ratio", or "custom". Default is "quantiles". If "custom" is selected, a `custom_rif_function` needs to be provided.
`weights`	numeric vector of non-negative observation weights, hence of same length as `dep_var`. The default (`NULL`) is equivalent to `weights = rep(1, length(dep_var))`.
`probs`	a vector of length 1 or more with probabilities of quantiles. Each quantile is indicated with a value between 0 and 1. Default is `c(1:9)/10`. If `statistic = "quantiles"`, a single RIF regression for every quantile in `probs` is estimated. An interquantile ratio (range) is defined by the ratio (difference) between the `max(probs)`-quantile and the `min(probs)`-quantile.
`custom_rif_function`	the RIF function to compute the RIF of the custom distributional statistic. Default is NULL. Only needs to be provided if `statistic = "custom"`. Every custom_rif_function needs the parameters `dep_var`, `weights` and `probs`. If they are not needed, they must be set to NULL in the function definition (e.g. `probs = NULL`). A custom function must return a data frame containing at least a "rif" and "weights" column. See `examples` for further details.
`na.action`	generic function that defines how NAs in the data should be handled. Default is `na.omit`, leading to exclusion of observations that contain one or more missings. See na.action for further details.
`bootstrap`	boolean (default = FALSE) indicating if bootstrapped standard errors will be computed
`bootstrap_iterations`	positive integer indicating the number of bootstrap iterations to execute. Only required if `bootstrap = TRUE`.
`cores`	positive integer indicating the number of cores to use when computing bootstrapped standard errors. Only required if `bootstrap = TRUE`.
`...`	additional parameters passed to the `custom_rif_function`. Apart from `dep_var`, `weights` and `probs` they must have a different name than the the ones in `rifreg`. For instance, if you want to pass a parameter `statistic` to the `custom_rif_function`, name it `custom_statistic`.

Value

rifreg returns an object of class "rifreg".

A "rifreg" object is a list containing the following components:

`estimates`	a matrix of RIF regression coefficients for each covariate and the intercept. In case of several quantiles, coefficient estimates for each quantile are provided. Equivalent to `coef()` call of an object of class `"lm"`.
`rif_lm`	one or several objects of class `"lm"`, containing the detailed RIF regression results.
`rif`	a data frame containing the RIF for each observation.
`bootstrap_se`	bootstrapped standard errors for each coefficient. Only provided if `bootstrap = TRUE`.
`bootstrap_vcov`	the bootstrapped variance-covariance matrix for each coefficient. Only provided if `bootstrap = TRUE`.
`statistic`	the distributional statistic for which the RIF was computed.
`custom_rif_function`	The custom RIF function in case it was provided.
`probs`	the probabilities of the quantiles that were computed, in case the distributional statistic requires quantiles.

References

Firpo, Sergio P., Nicole M. Fortin, and Thomas Lemieux. 2009. “Unconditional Quantile Regressions.” Econometrica 77(3): 953–73.

Examples


rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  weights = weights,
  probs = seq(0.1, 0.9, 0.1),
  bootstrap = FALSE
)


# custom function
custom_variance_function <- function(dep_var, weights, probs = NULL) {
  weighted_mean <- weighted.mean(x = dep_var, w = weights)
  rif <- (dep_var - weighted_mean)^2
  rif <- data.frame(rif, weights)
  names(rif) <- c("rif_variance", "weights")
  return(rif)
}

rifreg <- rifreg(
  formula = log(wage) ~ union + nonwhite + married + education + experience,
  data = men8385,
  statistic = "custom",
  weights = weights,
  probs = NULL,
  custom_rif_function = custom_variance_function,
  bootstrap = FALSE
)

rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  weights = weights,
  probs = seq(0.1, 0.9, 0.1),
  bootstrap = FALSE
)


# custom function
custom_variance_function <- function(dep_var, weights, probs = NULL) {
  weighted_mean <- weighted.mean(x = dep_var, w = weights)
  rif <- (dep_var - weighted_mean)^2
  rif <- data.frame(rif, weights)
  names(rif) <- c("rif_variance", "weights")
  return(rif)
}

rifreg <- rifreg(
  formula = log(wage) ~ union + nonwhite + married + education + experience,
  data = men8385,
  statistic = "custom",
  weights = weights,
  probs = NULL,
  custom_rif_function = custom_variance_function,
  bootstrap = FALSE
)

summary method for class "rifreg"

Description

summary method for class "rifreg"

Usage

## S3 method for class 'rifreg'
summary(object, vcov = sandwich::sandwich, ...)
## S3 method for class 'rifreg'
summary(object, vcov = sandwich::sandwich, ...)

Arguments

`object`	an object of class "rifreg", usually, a result of a call to rifreg.
`vcov`	Function to estimate covariance matrix of rifreg coefficients if covariance matrix has not been bootstrapped. Per default, heteroscedasticity-consistent (HC) standard errors are calculated using sandwich. Note: These standard errors do not take the variance introduced by estimating RIF into account.
`...`	other parameters to be passed to summary functions.

Value

the function summary.rifreg() returns a list of summary statistics derived from the rifreg object given in object. For further details see summary.lm.

Examples


rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  probs = seq(0.1, 0.9, 0.1),
  weights = weights
)
summary(rifreg)

rifreg <- rifreg(
  formula = log(wage) ~ union +
    nonwhite +
    married +
    education +
    experience,
  data = men8385,
  statistic = "quantiles",
  probs = seq(0.1, 0.9, 0.1),
  weights = weights
)
summary(rifreg)

Package 'rifreg'

Help Index

Check weights

Description

Usage

Arguments

Value

Examples

Generalized Lorenz ordinates

Description

Usage

Arguments

Value

Examples

Compute Gini coefficient

Description

Usage

Arguments

Value

References

Examples

Weighted ECDF value

Description

Usage

Arguments

Value

Examples

Estimate Recentered Influence Functions

Description

Usage

Arguments

Value

References

Examples

Estimate RIF of Gini coefficient

Description

Usage

Arguments

Value

References

Examples

Estimate RIF of interquantile range

Description

Usage

Arguments

Value

References

Examples

Estimate RIF of interquantile ratio

Description

Usage

Arguments

Value

References

Examples

Estimate RIF at the Mean

Description

Usage

Arguments

Value

Examples

Estimate RIF at Quantiles

Description

Usage

Arguments

Value

Functions

Examples

Estimate RIF of variance

Description

Usage

Arguments

Value

Examples

Integrate generalized Lorenz curve

Description

Usage

Arguments

Value

Examples

Plot the coefficients of a `rifreg` object