From f68f8b25f4e9898e954085f617e74557817cb0b4 Mon Sep 17 00:00:00 2001
From: Jaroslav Borodavka <df2994@kit.edu>
Date: Fri, 15 Mar 2024 13:24:00 +0100
Subject: [PATCH] Translated most of the code from German to English.
---
R/Comp_Statistics_Powertests_d_equal_2.R | 196 +++++++++++------------
1 file changed, 98 insertions(+), 98 deletions(-)
diff --git a/R/Comp_Statistics_Powertests_d_equal_2.R b/R/Comp_Statistics_Powertests_d_equal_2.R
index 4d7ee4a..86bf022 100644
--- a/R/Comp_Statistics_Powertests_d_equal_2.R
+++ b/R/Comp_Statistics_Powertests_d_equal_2.R
@@ -121,13 +121,13 @@ CircComp_stat_value <- function(data, index){
# test realizations for uniform distribution
CircComp_unif <- function(n, dim, index){
- return(list("Realisierung" = CircComp_stat_value(data = runif_sphere(n, dim),
+ return(list("realization" = CircComp_stat_value(data = runif_sphere(n, dim),
index = index)))
}
# alternatives according to paper by Cutting, Pandaveine und Verdebout ("TESTING UNIFORMITY ON HIGH-DIMENSIONAL SPHERES AGAINST MONOTONE ROTATIONALLY SYMMETRIC ALTERNATIVES")
CircComp_vMF <- function(n, dim, kappa, index){
- return(list("Realisierung" = CircComp_stat_value(data = rvmf(n, c(1,rep(0, dim-1)), kappa),
+ return(list("realization" = CircComp_stat_value(data = rvmf(n, c(1,rep(0, dim-1)), kappa),
index = index)))
}
@@ -135,41 +135,41 @@ CircComp_vMF <- function(n, dim, kappa, index){
CircComp_mixvmf_1 <- function(n, dim, p, index){
U = runif(n, 0, 1)
sample = (U<p)*rvmf(n, c(-1,rep(0,dim-1)), 1) + (U>=p)*rvmf(n, c(1,rep(0,dim-1)), 1)
- return(list("Realisierung" = CircComp_stat_value(data = sample, index = index)))
+ return(list("realization" = CircComp_stat_value(data = sample, index = index)))
}
CircComp_mixvmf_2 <- function(n, dim, p, index){
U = runif(n, 0, 1)
sample = (U<p)*rvmf(n, c(-1,rep(0,dim-1)), 1)+(U>=p)*rvmf(n, c(1,rep(0,dim-1)), 4)
- return(list("Realisierung" = CircComp_stat_value(data = sample, index = index)))
+ return(list("realization" = CircComp_stat_value(data = sample, index = index)))
}
CircComp_mixvmf_3 <- function(n, dim, index){
p = 0.25
U = runif(n, 0, 1)
sample = (U<p)*rvmf(n, rep(-1,dim), 2) + (U>=p)*(U<2*p)*rvmf(n, c(-1,rep(1,dim-1)), 3) + (U>=2*p)*rvmf(n, c(1,rep(0,dim-1)), 3)
- return(list("Realisierung" = CircComp_stat_value(data = sample, index = index)))
+ return(list("realization" = CircComp_stat_value(data = sample, index = index)))
}
CircComp_mixvmf_4 <- function(n, dim, index){
p = 1/3
U = runif(n, 0, 1)
sample = (U<p)*rvmf(n, rep(-1,dim), 2) + (U>=p)*(U<2*p)*rvmf(n, c(-1,rep(1,dim-1)), 3) + (U>=2*p)*rvmf(n, c(1,rep(0,dim-1)), 4)
- return(list("Realisierung" = CircComp_stat_value(data = sample, index = index)))
+ return(list("realization" = CircComp_stat_value(data = sample, index = index)))
}
CircComp_bing_1 <- function(n, dim, tau, index){
A = tau*diag(seq(1,dim))
- return(list("Realisierung" = CircComp_stat_value(data = rbingham(n, A), index = index)))
+ return(list("realization" = CircComp_stat_value(data = rbingham(n, A), index = index)))
}
CircComp_bing_2 <- function(n, dim, tau, index){
A = tau*diag(c(-dim, rep(0, dim-2), dim))
- return(list("Realisierung" = CircComp_stat_value(data = rbingham(n, A), index = index)))
+ return(list("realization" = CircComp_stat_value(data = rbingham(n, A), index = index)))
}
CircComp_LP <- function(n, dim, m, kappa, index){
- return(list("Realisierung" = CircComp_stat_value(data = rLP(n, m, c(1,rep(0,dim-1)), kappa), index = index)))
+ return(list("realization" = CircComp_stat_value(data = rLP(n, m, c(1,rep(0,dim-1)), kappa), index = index)))
}
#################################################################################################
@@ -202,17 +202,17 @@ DT_CircComp_unif <- setDT(DF_CircComp_unif)
# creating groups
DT_CircComp_unif[, grp := .GRP, by = c("n", "dim", "index")]
-DT_CircComp_unif = DT_CircComp_unif[order(grp, Realisierung)]
-DT_CircComp_unif_CA = DT_CircComp_unif[index == 5, list(n, "Dimension" = dim, "Index" = index, "Realisierung" = Realisierung,
- "Niveau_10" = lapply(.SD, quantile, probs = 0.10, na.rm = T),
- "Niveau_5" = lapply(.SD, quantile, probs = 0.05, na.rm = T),
- "Niveau_1" = lapply(.SD, quantile, probs = 0.01, na.rm = T)),
- by = grp, .SDcols = c("Realisierung")]
-DT_CircComp_unif = DT_CircComp_unif[index %in% c(seq(1,4), 6), list(n, "Dimension" = dim, "Index" = index, "Realisierung" = Realisierung,
- "Niveau_10" = lapply(.SD, quantile, probs = 0.90, na.rm = T),
- "Niveau_5" = lapply(.SD, quantile, probs = 0.95, na.rm = T),
- "Niveau_1" = lapply(.SD, quantile, probs = 0.99, na.rm = T)),
- by = grp, .SDcols = c("Realisierung")]
+DT_CircComp_unif = DT_CircComp_unif[order(grp, realization)]
+DT_CircComp_unif_CA = DT_CircComp_unif[index == 5, list(n, "dimension" = dim, "index" = index, "realization" = realization,
+ "level_10" = lapply(.SD, quantile, probs = 0.10, na.rm = T),
+ "level_5" = lapply(.SD, quantile, probs = 0.05, na.rm = T),
+ "level_1" = lapply(.SD, quantile, probs = 0.01, na.rm = T)),
+ by = grp, .SDcols = c("realization")]
+DT_CircComp_unif = DT_CircComp_unif[index %in% c(seq(1,4), 6), list(n, "dimension" = dim, "index" = index, "realization" = realization,
+ "level_10" = lapply(.SD, quantile, probs = 0.90, na.rm = T),
+ "level_5" = lapply(.SD, quantile, probs = 0.95, na.rm = T),
+ "level_1" = lapply(.SD, quantile, probs = 0.99, na.rm = T)),
+ by = grp, .SDcols = c("realization")]
DT_CircComp_unif = rbindlist(list(DT_CircComp_unif, DT_CircComp_unif_CA))
print(DT_CircComp_unif)
@@ -238,14 +238,14 @@ DT_CircComp_Test = na.omit(DT_CircComp_Test)
Power_CircComp = matrix(data = c(seq(1, grp_number), rep(0, 6*grp_number)), nrow = grp_number, ncol = 7)
for (k in 1:grp_number) {
- real = DT_CircComp_Test[grp==k]$Realisierung
+ real = DT_CircComp_Test[grp==k]$realization
n = DT_CircComp_Test[grp==k]$n[1]
dim = DT_CircComp_Test[grp==k]$dim[1]
index = DT_CircComp_Test[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[grp==k]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[grp==k]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[grp==k]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[grp==k]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[grp==k]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[grp==k]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -266,8 +266,8 @@ for (k in 1:grp_number) {
Power_CircComp[k, 7] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp = data.frame("Gruppe" = Power_CircComp[,1], "n" = Power_CircComp[,2], "dim" = Power_CircComp[,3], "Teststatistik" = Power_CircComp[,4],
- "Niveau_10" = Power_CircComp[,5], "Niveau_5" = Power_CircComp[,6], "Niveau_1" = Power_CircComp[,7])
+Power_CircComp = data.frame("group" = Power_CircComp[,1], "n" = Power_CircComp[,2], "dim" = Power_CircComp[,3], "test" = Power_CircComp[,4],
+ "level_10" = Power_CircComp[,5], "level_5" = Power_CircComp[,6], "level_1" = Power_CircComp[,7])
print(Power_CircComp)
#################################################################################################
@@ -294,15 +294,15 @@ DT_CircComp_vMF = na.omit(DT_CircComp_vMF)
Power_CircComp_vMF = matrix(data = c(seq(1,grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_vMF[grp==k]$Realisierung
- umfang = DT_CircComp_vMF[grp==k]$n[1]
+ real = DT_CircComp_vMF[grp==k]$realization
+ size = DT_CircComp_vMF[grp==k]$n[1]
dim = DT_CircComp_vMF[grp==k]$dim[1]
kappa = DT_CircComp_vMF[grp==k]$kappa[1]
index = DT_CircComp_vMF[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -315,7 +315,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_vMF[k, 2] = umfang
+ Power_CircComp_vMF[k, 2] = size
Power_CircComp_vMF[k, 3] = dim
Power_CircComp_vMF[k, 4] = kappa
Power_CircComp_vMF[k, 4] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", "Bingham")
@@ -324,9 +324,9 @@ for (k in 1:grp_number) {
Power_CircComp_vMF[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_vMF = data.frame("Gruppe" = Power_CircComp_vMF[,1], "n" = Power_CircComp_vMF[,2], "dim" = Power_CircComp_vMF[,3],
- "Kappa" = Power_CircComp_vMF[,4], "Teststatistik" = Power_CircComp_vMF[, 5], "Niveau_10" = Power_CircComp_vMF[,6],
- "Niveau_5" = Power_CircComp_vMF[,7], "Niveau_1" = Power_CircComp_vMF[,8])
+Power_CircComp_vMF = data.frame("group" = Power_CircComp_vMF[,1], "n" = Power_CircComp_vMF[,2], "dim" = Power_CircComp_vMF[,3],
+ "kappa" = Power_CircComp_vMF[,4], "test" = Power_CircComp_vMF[, 5], "level_10" = Power_CircComp_vMF[,6],
+ "level_5" = Power_CircComp_vMF[,7], "level_1" = Power_CircComp_vMF[,8])
print(Power_CircComp_vMF)
# Mix vMF with two centers (1)
@@ -350,15 +350,15 @@ DT_CircComp_mixvmf_1 = na.omit(DT_CircComp_mixvmf_1)
Power_CircComp_mixvmf_1 = matrix(data = c(seq(1,grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_mixvmf_1[grp==k]$Realisierung
- umfang = DT_CircComp_mixvmf_1[grp==k]$n[1]
+ real = DT_CircComp_mixvmf_1[grp==k]$realization
+ size = DT_CircComp_mixvmf_1[grp==k]$n[1]
dim = DT_CircComp_mixvmf_1[grp==k]$dim[1]
prob = DT_CircComp_mixvmf_1[grp==k]$p[1]
index = DT_CircComp_mixvmf_1[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -371,7 +371,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_mixvmf_1[k, 2] = umfang
+ Power_CircComp_mixvmf_1[k, 2] = size
Power_CircComp_mixvmf_1[k, 3] = dim
Power_CircComp_mixvmf_1[k, 4] = prob
Power_CircComp_mixvmf_1[k, 5] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", "Bingham")
@@ -380,9 +380,9 @@ for (k in 1:grp_number) {
Power_CircComp_mixvmf_1[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_mixvmf_1 = data.frame("Gruppe" = Power_CircComp_mixvmf_1[,1], "n" = Power_CircComp_mixvmf_1[,2], "dim" = Power_CircComp_mixvmf_1[,3],
- "Wahrscheinlichkeit" = Power_CircComp_mixvmf_1[,4], "Teststatistik" = Power_CircComp_mixvmf_1[,5], "Niveau_10" = Power_CircComp_mixvmf_1[,6],
- "Niveau_5" = Power_CircComp_mixvmf_1[,7], "Niveau_1" = Power_CircComp_mixvmf_1[,8])
+Power_CircComp_mixvmf_1 = data.frame("group" = Power_CircComp_mixvmf_1[,1], "n" = Power_CircComp_mixvmf_1[,2], "dim" = Power_CircComp_mixvmf_1[,3],
+ "Wahrscheinlichkeit" = Power_CircComp_mixvmf_1[,4], "test" = Power_CircComp_mixvmf_1[,5], "level_10" = Power_CircComp_mixvmf_1[,6],
+ "level_5" = Power_CircComp_mixvmf_1[,7], "level_1" = Power_CircComp_mixvmf_1[,8])
print(Power_CircComp_mixvmf_1)
# Mix vMF with two centers (2)
@@ -406,15 +406,15 @@ DT_CircComp_mixvmf_2 = na.omit(DT_CircComp_mixvmf_2)
Power_CircComp_mixvmf_2 = matrix(data = c(seq(1, grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_mixvmf_2[grp==k]$Realisierung
- umfang = DT_CircComp_mixvmf_2[grp==k]$n[1]
+ real = DT_CircComp_mixvmf_2[grp==k]$realization
+ size = DT_CircComp_mixvmf_2[grp==k]$n[1]
dim = DT_CircComp_mixvmf_2[grp==k]$dim[1]
prob = DT_CircComp_mixvmf_2[grp==k]$p[1]
index = DT_CircComp_mixvmf_2[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -427,7 +427,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_mixvmf_2[k, 2] = umfang
+ Power_CircComp_mixvmf_2[k, 2] = size
Power_CircComp_mixvmf_2[k, 3] = dim
Power_CircComp_mixvmf_2[k, 4] = prob
Power_CircComp_mixvmf_2[k, 5] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", "Bingham")
@@ -436,9 +436,9 @@ for (k in 1:grp_number) {
Power_CircComp_mixvmf_2[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_mixvmf_2 = data.frame("Gruppe" = Power_CircComp_mixvmf_2[,1], "n" = Power_CircComp_mixvmf_2[,2], "dim" = Power_CircComp_mixvmf_2[,3],
- "Wahrscheinlichkeit" = Power_CircComp_mixvmf_2[,4], "Teststatistik" = Power_CircComp_mixvmf_2[,5], "Niveau_10" = Power_CircComp_mixvmf_2[,6],
- "Niveau_5" = Power_CircComp_mixvmf_2[,7], "Niveau_1" = Power_CircComp_mixvmf_2[,8])
+Power_CircComp_mixvmf_2 = data.frame("group" = Power_CircComp_mixvmf_2[,1], "n" = Power_CircComp_mixvmf_2[,2], "dim" = Power_CircComp_mixvmf_2[,3],
+ "Wahrscheinlichkeit" = Power_CircComp_mixvmf_2[,4], "test" = Power_CircComp_mixvmf_2[,5], "level_10" = Power_CircComp_mixvmf_2[,6],
+ "level_5" = Power_CircComp_mixvmf_2[,7], "level_1" = Power_CircComp_mixvmf_2[,8])
print(Power_CircComp_mixvmf_2)
# Mix vMF with three centers (1)
@@ -461,14 +461,14 @@ DT_CircComp_mixvmf_3 = na.omit(DT_CircComp_mixvmf_3)
Power_CircComp_mixvmf_3 = matrix(data = c(seq(1,grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_mixvmf_3[grp==k]$Realisierung
- umfang = DT_CircComp_mixvmf_3[grp==k]$n[1]
+ real = DT_CircComp_mixvmf_3[grp==k]$realization
+ size = DT_CircComp_mixvmf_3[grp==k]$n[1]
dim = DT_CircComp_mixvmf_3[grp==k]$dim[1]
index = DT_CircComp_mixvmf_3[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -481,7 +481,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_mixvmf_3[k, 2] = umfang
+ Power_CircComp_mixvmf_3[k, 2] = size
Power_CircComp_mixvmf_3[k, 3] = dim
Power_CircComp_mixvmf_3[k, 4] = 0.25
Power_CircComp_mixvmf_3[k, 5] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", "Bingham")
@@ -490,9 +490,9 @@ for (k in 1:grp_number) {
Power_CircComp_mixvmf_3[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_mixvmf_3 = data.frame("Gruppe" = Power_CircComp_mixvmf_3[,1], "n" = Power_CircComp_mixvmf_3[,2], "dim" = Power_CircComp_mixvmf_3[,3],
- "Wahrscheinlichkeit" = Power_CircComp_mixvmf_3[,4], "Teststatistik" = Power_CircComp_mixvmf_3[, 5], "Niveau_10" = Power_CircComp_mixvmf_3[,6], "Niveau_5" = Power_CircComp_mixvmf_3[,7],
- "Niveau_1" = Power_CircComp_mixvmf_3[,8])
+Power_CircComp_mixvmf_3 = data.frame("group" = Power_CircComp_mixvmf_3[,1], "n" = Power_CircComp_mixvmf_3[,2], "dim" = Power_CircComp_mixvmf_3[,3],
+ "Wahrscheinlichkeit" = Power_CircComp_mixvmf_3[,4], "test" = Power_CircComp_mixvmf_3[, 5], "level_10" = Power_CircComp_mixvmf_3[,6], "level_5" = Power_CircComp_mixvmf_3[,7],
+ "level_1" = Power_CircComp_mixvmf_3[,8])
print(Power_CircComp_mixvmf_3)
# Mix vMF with three centers (2)
@@ -513,14 +513,14 @@ DT_CircComp_mixvmf_4 = na.omit(DT_CircComp_mixvmf_4)
Power_CircComp_mixvmf_4 = matrix(data = c(seq(1,grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_mixvmf_4[grp==k]$Realisierung
- umfang = DT_CircComp_mixvmf_4[grp==k]$n[1]
+ real = DT_CircComp_mixvmf_4[grp==k]$realization
+ size = DT_CircComp_mixvmf_4[grp==k]$n[1]
dim = DT_CircComp_mixvmf_4[grp==k]$dim[1]
index = DT_CircComp_mixvmf_4[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -533,7 +533,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_mixvmf_4[k, 2] = umfang
+ Power_CircComp_mixvmf_4[k, 2] = size
Power_CircComp_mixvmf_4[k, 3] = dim
Power_CircComp_mixvmf_4[k, 4] = round(1/3, digits = 2)
Power_CircComp_mixvmf_4[k, 5] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", , "Bingham")
@@ -542,9 +542,9 @@ for (k in 1:grp_number) {
Power_CircComp_mixvmf_4[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_mixvmf_4 = data.frame("Gruppe" = Power_CircComp_mixvmf_4[,1], "n" = Power_CircComp_mixvmf_4[,2], "dim" = Power_CircComp_mixvmf_4[,3],
- "Wahrscheinlichkeit" = Power_CircComp_mixvmf_4[,4], "Teststatistik" = Power_CircComp_mixvmf_4[,5], "Niveau_10" = Power_CircComp_mixvmf_4[,6], "Niveau_5" = Power_CircComp_mixvmf_4[,7],
- "Niveau_1" = Power_CircComp_mixvmf_4[,8])
+Power_CircComp_mixvmf_4 = data.frame("group" = Power_CircComp_mixvmf_4[,1], "n" = Power_CircComp_mixvmf_4[,2], "dim" = Power_CircComp_mixvmf_4[,3],
+ "Wahrscheinlichkeit" = Power_CircComp_mixvmf_4[,4], "test" = Power_CircComp_mixvmf_4[,5], "level_10" = Power_CircComp_mixvmf_4[,6], "level_5" = Power_CircComp_mixvmf_4[,7],
+ "level_1" = Power_CircComp_mixvmf_4[,8])
print(Power_CircComp_mixvmf_4)
@@ -569,15 +569,15 @@ DT_CircComp_bing_1 = na.omit(DT_CircComp_bing_1)
Power_CircComp_bing_1 = matrix(data = c(seq(1,grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_bing_1[grp==k]$Realisierung
- umfang = DT_CircComp_bing_1[grp==k]$n[1]
+ real = DT_CircComp_bing_1[grp==k]$realization
+ size = DT_CircComp_bing_1[grp==k]$n[1]
dim = DT_CircComp_bing_1[grp==k]$dim[1]
tau = DT_CircComp_bing_1[grp==k]$tau[1]
index = DT_CircComp_bing_1[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -590,7 +590,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_bing_1[k, 2] = umfang
+ Power_CircComp_bing_1[k, 2] = size
Power_CircComp_bing_1[k, 3] = dim
Power_CircComp_bing_1[k, 4] = tau
Power_CircComp_bing_1[k, 5] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", "Bingham")
@@ -599,9 +599,9 @@ for (k in 1:grp_number) {
Power_CircComp_bing_1[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_bing_1 = data.frame("Gruppe" = Power_CircComp_bing_1[,1], "n" = Power_CircComp_bing_1[,2], "dim" = Power_CircComp_bing_1[,3],
- "Tau" = Power_CircComp_bing_1[,4], "Teststatistik" = Power_CircComp_bing_1[,5], "Niveau_10" = Power_CircComp_bing_1[,6],
- "Niveau_5" = Power_CircComp_bing_1[,7], "Niveau_1" = Power_CircComp_bing_1[,8])
+Power_CircComp_bing_1 = data.frame("group" = Power_CircComp_bing_1[,1], "n" = Power_CircComp_bing_1[,2], "dim" = Power_CircComp_bing_1[,3],
+ "tau" = Power_CircComp_bing_1[,4], "test" = Power_CircComp_bing_1[,5], "level_10" = Power_CircComp_bing_1[,6],
+ "level_5" = Power_CircComp_bing_1[,7], "level_1" = Power_CircComp_bing_1[,8])
print(Power_CircComp_bing_1)
# Bingham (2)
@@ -625,15 +625,15 @@ DT_CircComp_bing_2 = na.omit(DT_CircComp_bing_2)
Power_CircComp_bing_2 = matrix(data = c(seq(1,grp_number), rep(0, 7*grp_number)), nrow = grp_number, ncol = 8)
for (k in 1:grp_number) {
- real = DT_CircComp_bing_2[grp==k]$Realisierung
- umfang = DT_CircComp_bing_2[grp==k]$n[1]
+ real = DT_CircComp_bing_2[grp==k]$realization
+ size = DT_CircComp_bing_2[grp==k]$n[1]
dim = DT_CircComp_bing_2[grp==k]$dim[1]
tau = DT_CircComp_bing_2[grp==k]$tau[1]
index = DT_CircComp_bing_2[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -646,7 +646,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_bing_2[k, 2] = umfang
+ Power_CircComp_bing_2[k, 2] = size
Power_CircComp_bing_2[k, 3] = dim
Power_CircComp_bing_2[k, 4] = tau
Power_CircComp_bing_2[k, 5] = switch(index, "Kuiper", "Watson", "Ajne", "Rayleigh", "Cuesta-Albertos", "Bingham")
@@ -655,9 +655,9 @@ for (k in 1:grp_number) {
Power_CircComp_bing_2[k, 8] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_bing_2 = data.frame("Gruppe" = Power_CircComp_bing_2[,1], "n" = Power_CircComp_bing_2[,2], "dim" = Power_CircComp_bing_2[,3],
- "Tau" = Power_CircComp_bing_2[,4], "Teststatistik" = Power_CircComp_bing_2[,5], "Niveau_10" = Power_CircComp_bing_2[,6],
- "Niveau_5" = Power_CircComp_bing_2[,7], "Niveau_1" = Power_CircComp_bing_2[,8])
+Power_CircComp_bing_2 = data.frame("group" = Power_CircComp_bing_2[,1], "n" = Power_CircComp_bing_2[,2], "dim" = Power_CircComp_bing_2[,3],
+ "tau" = Power_CircComp_bing_2[,4], "test" = Power_CircComp_bing_2[,5], "level_10" = Power_CircComp_bing_2[,6],
+ "level_5" = Power_CircComp_bing_2[,7], "level_1" = Power_CircComp_bing_2[,8])
print(Power_CircComp_bing_2)
# Legendre polynomial
@@ -682,16 +682,16 @@ DT_CircComp_LP = na.omit(DT_CircComp_LP)
Power_CircComp_LP = matrix(data = c(seq(1,grp_number), rep(0, 8*grp_number)), nrow = grp_number, ncol = 9)
for (k in 1:grp_number) {
- real = DT_CircComp_LP[grp==k]$Realisierung
- umfang = DT_CircComp_LP[grp==k]$n[1]
+ real = DT_CircComp_LP[grp==k]$realization
+ size = DT_CircComp_LP[grp==k]$n[1]
dim = DT_CircComp_LP[grp==k]$dim[1]
m = DT_CircComp_LP[grp==k]$m[1]
kappa = DT_CircComp_LP[grp==k]$kappa[1]
index = DT_CircComp_LP[grp==k]$index[1]
- niv_10 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_10[[1]]
- niv_5 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_5[[1]]
- niv_1 = DT_CircComp_unif[n == umfang & Dimension == dim & Index == index]$Niveau_1[[1]]
+ niv_10 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_10[[1]]
+ niv_5 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_5[[1]]
+ niv_1 = DT_CircComp_unif[n == size & dimension == dim & index == index]$level_1[[1]]
if (index == 5){
dec_10 = sapply(real, Decision_CA, critical = niv_10)
@@ -704,7 +704,7 @@ for (k in 1:grp_number) {
dec_1 = sapply(real, Decision, critical = niv_1)
}
- Power_CircComp_LP[k, 2] = umfang
+ Power_CircComp_LP[k, 2] = size
Power_CircComp_LP[k, 3] = dim
Power_CircComp_LP[k, 4] = m
Power_CircComp_LP[k, 5] = kappa
@@ -714,9 +714,9 @@ for (k in 1:grp_number) {
Power_CircComp_LP[k, 9] = sum(dec_1[dec_1==1])/length(dec_1)
}
-Power_CircComp_LP = data.frame("Gruppe" = Power_CircComp_LP[,1], "n" = Power_CircComp_LP[,2], "dim" = Power_CircComp_LP[,3],
- "m" = Power_CircComp_LP[,4], "Kappa" = Power_CircComp_LP[,5], "Teststatistik" = Power_CircComp_LP[,6], "Niveau_10" = Power_CircComp_LP[,7],
- "Niveau_5" = Power_CircComp_LP[,8], "Niveau_1" = Power_CircComp_LP[,9])
+Power_CircComp_LP = data.frame("group" = Power_CircComp_LP[,1], "n" = Power_CircComp_LP[,2], "dim" = Power_CircComp_LP[,3],
+ "m" = Power_CircComp_LP[,4], "kappa" = Power_CircComp_LP[,5], "test" = Power_CircComp_LP[,6], "level_10" = Power_CircComp_LP[,7],
+ "level_5" = Power_CircComp_LP[,8], "level_1" = Power_CircComp_LP[,9])
print(Power_CircComp_LP)
save.image(file = "Comp_Statistics_Powertests_d_equal_2_Daten.RData")
--
GitLab