tests/tests.R
In matthewclegg/egcm: Engle-Granger Cointegration Models
# egcm.R
# Copyright (C) 2014 by Matthew Clegg
# Tests for the Engle-Granger cointegration models package
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# A copy of the GNU General Public License is available at
# http://www.r-project.org/Licenses/
test <- function(expr, out="") {
# expr is a string representing an R expression, and
# out is the output that is expected. Prints and evaluates
# expr. If out is given and it matches the output of
# evaluating expr, returns TRUE. Otherwise, returns FALSE.
cat("\n", expr, "\n", sep="")
if (!missing(out)) cat("Expecting:",out, "", sep="\n")
val <- eval(parse(text=expr))
cat("Result:\n")
print(val)
if (!missing(out) &&
all(as.character(val) != as.character(out))) {
return(FALSE)
}
return(TRUE)
}
assert <- function (expr, out) {
# expr is astring representing an R expression,
# and out is the output that is expected. Prints
# and evaluates expr. If out matches the output of
# evaluating expr, returns TRUE. Otherwise, stops
# the execution with an error message.
if (!test(expr, out)) {
stop("Expression does not evaluate to its expected value\n")
}
}
egcm_tests <- function (include.inexact = FALSE) {
# If include.inexact is TRUE, then includes those tests for
# which an exact answer cannot be given.
if (include.inexact) {
test("ur_power(pp.test, nrep=100)", 0.4)
test("ur_power_table(pp.test, nrep=10)")
test("egcm:::ur_specificity(pp.test, function() cumsum(rnorm(100)))",
paste("0.01 0.02 0.05 0.1 0.2 0.5",
" 0.01 0.02 0.05 0.1 0.2 0.5", sep="\n"))
test("adf_power(nrep=100)", 0.26)
test("adf_power_table(nrep=10)")
}
assert("bvr_rho(1:100)", 0.10001)
assert("bvr.test(1:100+sin(1:100))$p.value", 0.999)
assert("bvr.test(1:100+sin(1:100),TRUE)$p.value", 0.001)
if (include.inexact) {
test("bvr_power(nrep=100)", 0.36)
test("bvr_power_table(nrep=10)")
}
if (include.inexact) {
test("pgff_rho_ws(1:100)", 1.019777)
test("pgff_rho_ws(1:100 + sin(1:100))", 1.018697)
test("pgff_rho_ws(1:100 + sin(1:100), TRUE)", 0.5464542)
}
assert("pgff.test(1:100+sin(1:100))$p.value > 0.99", TRUE)
assert("pgff.test(1:100+sin(1:100),TRUE)$p.value", 0.001)
if (include.inexact) {
test("pgff_power(nrep=100)", 0.75)
test("pgff_power_table(nrep=10)")
}
assert("length(egcm.urtests())", 11)
assert("egcm.default.urtest()", "pp")
assert("egcm.set.default.urtest(\"adf\"); egcm.default.urtest()",
"adf")
assert("egcm.set.default.urtest(\"pgff\"); egcm.default.urtest()",
"pgff")
assert("length(egcm.i1tests())", 4)
assert("egcm.default.i1test()", "pp")
assert("egcm.set.default.i1test(\"adf\"); egcm.default.i1test()",
"adf")
assert("egcm.set.default.i1test(\"pgff\"); egcm.default.i1test()",
"pgff")
assert("egcm.default.pvalue()", 0.05)
assert("egcm.set.default.pvalue(0.01); egcm.default.pvalue()", 0.01)
assert("egcm.set.default.pvalue(0.05); egcm.default.pvalue()", 0.05)
assert("detrend(1:5)", rep(0,5))
test("egcm(1:100, 1:100)",
"Y[i] = 1.0000 X[i] - 0.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm(1:100, (1:100)*2)",
"Y[i] = 2.0000 X[i] - 0.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm(1:100, (1:100)*2, normalize=TRUE)",
"Y[i] = 1.0000 X[i] - 0.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm (1:100, (1:100)*2+3)",
"Y[i] = 2.0000 X[i] + 3.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm (1:100, (1:100)*2+3, include.const=FALSE)",
"Y[i] = 2.0448 X[i] + 0.0000 + R[i], R[i] = 1.0000 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0026) (0.0000) (0.0026)
R[100] = -1.4776 (t = -1.137)")
test("egcm(1:100, 1:100 + sin(1:100))",
"Y[i] = 0.9988 X[i] + 0.0583 + R[i], R[i] = 0.5901 R[i-1] + eps[i], eps ~ N(0, 0.5957^2)
(0.0025) (0.3309) (0.0842)
R[100] = -0.4467 (t = -0.628)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:100, 1:100 + sin(1:100), debias=FALSE)",
"Y[i] = 0.9988 X[i] + 0.0583 + R[i], R[i] = 0.5452 R[i-1] + eps[i], eps ~ N(0, 0.5949^2)
(0.0025) (0.3305) (0.0842)
R[100] = -0.4467 (t = -0.628)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:101, c(1:100 + sin(1:100), 110), robust=FALSE)",
"Y[i] = 1.0041 X[i] - 0.1211 + R[i], R[i] = 0.4712 R[i-1] + eps[i], eps ~ N(0, 1.0925^2)
(0.0039) (0.5923) (0.1498)
R[101] = 8.7073 (t = 7.665)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:101, c(1:100 + sin(1:100), 110), robust=TRUE)",
"Y[i] = 0.9997 X[i] + 0.0284 + R[i], R[i] = 0.5770 R[i-1] + eps[i], eps ~ N(0, 1.1034^2)
(0.0025) (0.5710) (0.0861)
R[101] = 9.0019 (t = 7.874)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:25, 1:25+sin(1:25))",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: X and Y do not appear to be cointegrated.")
test("egcm(exp(1:25), exp(1:25+sin(1:25)), log=TRUE)",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: X and Y do not appear to be cointegrated.")
test("egcm(1:25, 1:25+sin(1:25), urtest=\"adf\")",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:25, 1:25+sin(1:25), p.value=0.2)",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: X and Y do not appear to be cointegrated.")
test("try(egcm(1:100, c(1:49,NA,51:100)))",
"Y[i] = 1.0000 X[i] - 0.0000 + R[i], R[i] = -0.0757 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.1004)
R[99] = 0.0000 (t = 0.001)")
test("try(egcm(1:100, c(1:49,NA,51:100), na.fail), silent=TRUE)",
"[1] \"Error in na.fail.default(S12) : missing values in object\\n\"
attr(,\"class\")
[1] \"try-error\"
attr(,\"condition\")
<simpleError in na.fail.default(S12): missing values in object>")
assert("is.cointegrated(egcm(1:100, 1:100))", TRUE)
assert("is.cointegrated(egcm(1:100, (1:100)^2))", FALSE)
assert("is.ar1(egcm(1:100, 1:100))", TRUE)
assert("is.ar1(egcm(1:100, 1:100 + sin(1:100)))", FALSE)
assert("acor(1:10)", 1)
assert("acor(c(1,0,1,0,1,0))", -1)
assert("acor(c(1,0,-1,0,1,0))", 0)
if (include.inexact) {
test("sum(replicate(100, pgff.test(rar1(100, a1=0.5))$p.value < 0.05))", 100)
test("sum(replicate(100, pgff.test(rar1(100, a1=1))$p.value < 0.05))", 5)
}
# Due to differences in machine precision, we cannot
# test these exactly, but at least we can try different
# combinations of parameters and see if the function
# at least executes.
}
library(egcm)
egcm_tests()
matthewclegg/egcm documentation built on March 5, 2023, 6:33 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# egcm.R
# Copyright (C) 2014 by Matthew Clegg
# Tests for the Engle-Granger cointegration models package
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# A copy of the GNU General Public License is available at
# http://www.r-project.org/Licenses/
test <- function(expr, out="") {
# expr is a string representing an R expression, and
# out is the output that is expected. Prints and evaluates
# expr. If out is given and it matches the output of
# evaluating expr, returns TRUE. Otherwise, returns FALSE.
cat("\n", expr, "\n", sep="")
if (!missing(out)) cat("Expecting:",out, "", sep="\n")
val <- eval(parse(text=expr))
cat("Result:\n")
print(val)
if (!missing(out) &&
all(as.character(val) != as.character(out))) {
return(FALSE)
}
return(TRUE)
}
assert <- function (expr, out) {
# expr is astring representing an R expression,
# and out is the output that is expected. Prints
# and evaluates expr. If out matches the output of
# evaluating expr, returns TRUE. Otherwise, stops
# the execution with an error message.
if (!test(expr, out)) {
stop("Expression does not evaluate to its expected value\n")
}
}
egcm_tests <- function (include.inexact = FALSE) {
# If include.inexact is TRUE, then includes those tests for
# which an exact answer cannot be given.
if (include.inexact) {
test("ur_power(pp.test, nrep=100)", 0.4)
test("ur_power_table(pp.test, nrep=10)")
test("egcm:::ur_specificity(pp.test, function() cumsum(rnorm(100)))",
paste("0.01 0.02 0.05 0.1 0.2 0.5",
" 0.01 0.02 0.05 0.1 0.2 0.5", sep="\n"))
test("adf_power(nrep=100)", 0.26)
test("adf_power_table(nrep=10)")
}
assert("bvr_rho(1:100)", 0.10001)
assert("bvr.test(1:100+sin(1:100))$p.value", 0.999)
assert("bvr.test(1:100+sin(1:100),TRUE)$p.value", 0.001)
if (include.inexact) {
test("bvr_power(nrep=100)", 0.36)
test("bvr_power_table(nrep=10)")
}
if (include.inexact) {
test("pgff_rho_ws(1:100)", 1.019777)
test("pgff_rho_ws(1:100 + sin(1:100))", 1.018697)
test("pgff_rho_ws(1:100 + sin(1:100), TRUE)", 0.5464542)
}
assert("pgff.test(1:100+sin(1:100))$p.value > 0.99", TRUE)
assert("pgff.test(1:100+sin(1:100),TRUE)$p.value", 0.001)
if (include.inexact) {
test("pgff_power(nrep=100)", 0.75)
test("pgff_power_table(nrep=10)")
}
assert("length(egcm.urtests())", 11)
assert("egcm.default.urtest()", "pp")
assert("egcm.set.default.urtest(\"adf\"); egcm.default.urtest()",
"adf")
assert("egcm.set.default.urtest(\"pgff\"); egcm.default.urtest()",
"pgff")
assert("length(egcm.i1tests())", 4)
assert("egcm.default.i1test()", "pp")
assert("egcm.set.default.i1test(\"adf\"); egcm.default.i1test()",
"adf")
assert("egcm.set.default.i1test(\"pgff\"); egcm.default.i1test()",
"pgff")
assert("egcm.default.pvalue()", 0.05)
assert("egcm.set.default.pvalue(0.01); egcm.default.pvalue()", 0.01)
assert("egcm.set.default.pvalue(0.05); egcm.default.pvalue()", 0.05)
assert("detrend(1:5)", rep(0,5))
test("egcm(1:100, 1:100)",
"Y[i] = 1.0000 X[i] - 0.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm(1:100, (1:100)*2)",
"Y[i] = 2.0000 X[i] - 0.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm(1:100, (1:100)*2, normalize=TRUE)",
"Y[i] = 1.0000 X[i] - 0.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm (1:100, (1:100)*2+3)",
"Y[i] = 2.0000 X[i] + 3.0000 + R[i], R[i] = -0.1241 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.0992)
R[100] = -0.0000 (t = -0.089)")
test("egcm (1:100, (1:100)*2+3, include.const=FALSE)",
"Y[i] = 2.0448 X[i] + 0.0000 + R[i], R[i] = 1.0000 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0026) (0.0000) (0.0026)
R[100] = -1.4776 (t = -1.137)")
test("egcm(1:100, 1:100 + sin(1:100))",
"Y[i] = 0.9988 X[i] + 0.0583 + R[i], R[i] = 0.5901 R[i-1] + eps[i], eps ~ N(0, 0.5957^2)
(0.0025) (0.3309) (0.0842)
R[100] = -0.4467 (t = -0.628)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:100, 1:100 + sin(1:100), debias=FALSE)",
"Y[i] = 0.9988 X[i] + 0.0583 + R[i], R[i] = 0.5452 R[i-1] + eps[i], eps ~ N(0, 0.5949^2)
(0.0025) (0.3305) (0.0842)
R[100] = -0.4467 (t = -0.628)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:101, c(1:100 + sin(1:100), 110), robust=FALSE)",
"Y[i] = 1.0041 X[i] - 0.1211 + R[i], R[i] = 0.4712 R[i-1] + eps[i], eps ~ N(0, 1.0925^2)
(0.0039) (0.5923) (0.1498)
R[101] = 8.7073 (t = 7.665)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:101, c(1:100 + sin(1:100), 110), robust=TRUE)",
"Y[i] = 0.9997 X[i] + 0.0284 + R[i], R[i] = 0.5770 R[i-1] + eps[i], eps ~ N(0, 1.1034^2)
(0.0025) (0.5710) (0.0861)
R[101] = 9.0019 (t = 7.874)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:25, 1:25+sin(1:25))",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: X and Y do not appear to be cointegrated.")
test("egcm(exp(1:25), exp(1:25+sin(1:25)), log=TRUE)",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: X and Y do not appear to be cointegrated.")
test("egcm(1:25, 1:25+sin(1:25), urtest=\"adf\")",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: The series seem cointegrated but the residuals are not AR(1).")
test("egcm(1:25, 1:25+sin(1:25), p.value=0.2)",
"Y[i] = 0.9818 X[i] + 0.2349 + R[i], R[i] = 0.7471 R[i-1] + eps[i], eps ~ N(0, 0.6132^2)
(0.0201) (0.4284) (0.1711)
R[25] = 0.0890 (t = 0.125)
WARNING: X and Y do not appear to be cointegrated.")
test("try(egcm(1:100, c(1:49,NA,51:100)))",
"Y[i] = 1.0000 X[i] - 0.0000 + R[i], R[i] = -0.0757 R[i-1] + eps[i], eps ~ N(0, 0.0000^2)
(0.0000) (0.0000) (0.1004)
R[99] = 0.0000 (t = 0.001)")
test("try(egcm(1:100, c(1:49,NA,51:100), na.fail), silent=TRUE)",
"[1] \"Error in na.fail.default(S12) : missing values in object\\n\"
attr(,\"class\")
[1] \"try-error\"
attr(,\"condition\")
<simpleError in na.fail.default(S12): missing values in object>")
assert("is.cointegrated(egcm(1:100, 1:100))", TRUE)
assert("is.cointegrated(egcm(1:100, (1:100)^2))", FALSE)
assert("is.ar1(egcm(1:100, 1:100))", TRUE)
assert("is.ar1(egcm(1:100, 1:100 + sin(1:100)))", FALSE)
assert("acor(1:10)", 1)
assert("acor(c(1,0,1,0,1,0))", -1)
assert("acor(c(1,0,-1,0,1,0))", 0)
if (include.inexact) {
test("sum(replicate(100, pgff.test(rar1(100, a1=0.5))$p.value < 0.05))", 100)
test("sum(replicate(100, pgff.test(rar1(100, a1=1))$p.value < 0.05))", 5)
}
# Due to differences in machine precision, we cannot
# test these exactly, but at least we can try different
# combinations of parameters and see if the function
# at least executes.
}
library(egcm)
egcm_tests()
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.