| Title: | Estimating Consistent Tests for Stochastic Dominance |
| Version: | 0.0.1 |
| Description: | Stochastic dominance tests help ranking different distributions. The package implements the consistent test for stochastic dominance by Barrett and Donald (2003) <doi:10.1111/1468-0262.00390>. Specifically, it implements Barrett and Donald's Kolmogorov-Smirnov type tests for first- and second-order stochastic dominance based on bootstrapping 2 and 1. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.2.3 |
| Imports: | dplyr, tibble, ggplot2, pracma, tidyr |
| Suggests: | testthat |
| NeedsCompilation: | no |
| Packaged: | 2024-01-31 11:58:52 UTC; f80866559 |
| Author: | Sergei Schaub 
[aut, cre],
Agroscope [cph] |
| Maintainer: | Sergei Schaub <sergei.schaub@agroscope.admin.ch> |
| Repository: | CRAN |
| Date/Publication: | 2024-02-02 12:40:02 UTC |
plot difference ecdfs
Description
This function computes the values of the cumulative difference of two empirical cumulative distribution function and plots the values.
Usage
dif_ecdf_plot(data_1, data_2, bins_size)
Arguments
data_1 |
data 1. |
data_2 |
data 2. |
bins_size |
bin size. |
Details
This function computes the values of the cumulative difference of two empirical cumulative distribution function and plots the values. This relates two showing second-order stochastic dominance.
Value
The function returns a plot as a ggplot2 object.
Examples
# load stodom
require(stodom)
data_a <- rnorm(500, 3, 2)
data_b <- rnorm(500, 1, 2)
# plot cumulative difference between two ecdfs
dif_ecdf_plot(data_1 = data_a, data_2 = data_b, bins_size = 0.1)
values of two ecdf and their cumulative difference
Description
This function computes the values of two empirical cumulative distribution function as well as their cumulative differences.
Usage
ecdf_dat_g(data_1, data_2, bins_size)
Arguments
data_1 |
data 1. |
data_2 |
data 2. |
bins_size |
bin size. |
Details
This function computes the values of two empirical cumulative distribution function as well as their cumulative differences.
Value
The function returns a data table.
Examples
# load stodom
require(stodom)
data_a <- rnorm(500, 3, 2)
data_b <- rnorm(500, 1, 2)
# compute the values of two ecdfs and their cumulative differences.
ecdf_dat_g(data_1 = data_a, data_2 = data_b, bins_size = 1)
plot ecdfs
Description
This function computes the values of two empirical cumulative distribution function and plots the values.
Usage
ecdf_plot(data_1, data_2, bins_size)
Arguments
data_1 |
data 1. |
data_2 |
data 2. |
bins_size |
bin size. |
Details
This function computes the values of two empirical cumulative distribution function and plots the values.
Value
The function returns a plot as a ggplot2 object.
Examples
# load stodom
require(stodom)
data_a <- rnorm(500, 3, 2)
data_b <- rnorm(500, 1, 2)
# plot ecdfs
ecdf_plot(data_1 = data_a, data_2 = data_b, bins_size = 0.1)
first-order stochastic dominance test
Description
This function tests for first-order stochastic dominance.
Usage
fo_stodom(data_1, data_2, bins_size, n_draws, useed, variable_1, variable_2, type)
Arguments
data_1 |
data 1. |
data_2 |
data 2. |
bins_size |
bin size. |
n_draws |
number of draws to compute p values (default = 500). |
useed |
user defined seed |
variable_1 |
name of a (as a string); only for the output table (default = "a"). |
variable_2 |
name of b (as a string); only for the output table (default = "b"). |
type |
type of bootstrapped test, bootstrapping 1 and 2 of Barrett and Donald (2003) are available (default = "boot2"). |
Details
This function computes the consistent test of first-order stochastic dominance following Barrett and Donald (2003). In detail, this function estimate their Kolmogorov-Smirnov type tests based on bootstrapping 2. The function was implemented as part of Schaub xxx
Value
The function returns a list object containing the p-values of two dominance tests (i.e., variable 1 vs. variable 1 and variable 2 vs. variable 1).
References
Barrett, G. F., & Donald, S. G. (2003). Consistent tests for stochastic dominance. Econometrica, 71(1), 71-104.
Schaub, S. & El Benni, N. (2024). How do price (risk) changes influence farmers’ preference to reduce fertilizer application?
Examples
# load stodom
require(stodom)
data_a <- rnorm(500, 3, 2)
data_b <- rnorm(500, 1, 2)
# estimate first-order stochastic dominance
fo_stodom(data_1 = data_a, data_2 = data_b, n_draws = 100, useed = 1, bins_size = 1)
second-order stochastic dominance test
Description
This function tests for second-order stochastic dominance.
Usage
so_stodom(data_1, data_2, bins_size, n_draws, useed, variable_1, variable_2, type)
Arguments
data_1 |
data 1. |
data_2 |
data 2. |
bins_size |
bin size. |
n_draws |
number of draws to compute p values (default = 500). |
useed |
user defined seed |
variable_1 |
name of a (as a string); only for the output table (default = "a"). |
variable_2 |
name of b (as a string); only for the output table (default = "b"). |
type |
type of bootstrapped test, bootstrapping 1 and 2 of Barrett and Donald (2003) are available (default = "boot2"). |
Details
This function computes the consistent test of second-order stochastic dominance following Barrett and Donald (2003). In detail, this function estimate their Kolmogorov-Smirnov type tests based on bootstrapping 2. The function was implemented as part of Schaub xxx
Value
The function returns a list object containing the p-values of two dominance tests (i.e., variable 1 vs. variable 1 and variable 2 vs. variable 1).
References
Barrett, G. F., & Donald, S. G. (2003). Consistent tests for stochastic dominance. Econometrica, 71(1), 71-104.
Schaub, S. & El Benni, N. (2024). How do price (risk) changes influence farmers’ preference to reduce fertilizer application?
Examples
# load stodom
require(stodom)
data_a <- rnorm(500, 3, 2)
data_b <- rnorm(500, 1, 2)
# estimate second-order stochastic dominance
so_stodom(data_1 = data_a, data_2 = data_b, n_draws = 100, useed = 1, bins_size = 1)