datenz
surv_obj <- Surv(time = datenr, event = datenz)
# Kaplan-Meier-Schätzer berechnen
km_fit <- survfit(Surv(datenr, datenz) ~ 1, data=df)
km_fit
load("o:/eckhard/Forschung/SPC/alabsi/daten/TEP_Faulty_Training.rdata")
plot(c(1,4,5,6,7,3,2,1))
plot(faulty_training[1:500])
plot(faulty_training$xmeas_1[1:500])
plot(faulty_training$xmeas_1[1:500],type = "l")
plot(faulty_training$xmeas_1[1:5000],type = "l")
view(faulty_training$faultNumber[1:1000])
View(faulty_training$faultNumber[1:1000])
View(faulty_training)
which(faulty_training$faultNumber[1:1000]>1)
shiny::runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
mf$vgrid
dat()$v
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
mf$ggrid
shiny::runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
input$distr
fit_obj()
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
fito <- fit_obj()
pfad_dat<- "o:/eckhard/stat/R/zuverl/deskr_stat/"
datname<-paste0(pfad_dat,"aluminiumbl.csv")
daten<- read.csv2(datname, header=T,skip=0,dec=".")
fitres<- fitdistrplus::fitdist(daten, "gamma", method="mle")
daten
fitres<- fitdistrplus::fitdist(daten, "gamma", method="mle")
length(daten)
daten[1]
str(daten)
pfad_dat<- "o:/eckhard/stat/R/zuverl/deskr_stat/"
datname<-paste0(pfad_dat,"aluminiumbl.csv")
daten<- unname(read.csv2(datname, header=T,skip=0,dec="."))
fitres<- fitdistrplus::fitdist(daten, "gamma", method="mle")
daten
daten[,1]
fitres<- fitdistrplus::fitdist(daten[,1], "gamma", method="mle")
sh <- unname(fitres$estimate["shape"])
sc <- unname(fitres$estimate["scale"])
res <- ADGofTest::ad.test(daten, pgamma, shape = sh, scale = sc)
res <- ADGofTest::ad.test(daten[,1], pgamma, shape = sh, scale = sc)
sh
sc
fitres$estimate["scale"]
fitres$estimate
sc <- unname(fitres$estimate["rate"])
res <- ADGofTest::ad.test(daten[,1], pgamma, shape = sh, rate = sc)
res
runApp('Eckhard/stat/R/zuverl/deskr_stat')
fit_obj()
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
library(weibullness)
pfad_dat<- "o:/eckhard/stat/R/zuverl/deskr_stat/"
datname<-paste0(pfad_dat,"aluminiumbl.csv")
fitres<- weibull.mle(daten)
daten<- unname(read.csv2(datname, header=T,skip=0,dec="."))
fitres<- weibull.mle(daten)
fitres<- weibull.mle(daten,a=0.5)
daten
fitres<- weibull.mle(daten[,1],a=0.5)
fitres
fitres<- weibull.mle(daten[,1],a=0.9)
fitres
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
ppoints
ppoints(10,0.5)
ppoints(10,0.9)
pfad_dat<- "o:/eckhard/stat/R/zuverl/deskr_stat/"
datname<-paste0(pfad_dat,"aluminiumbl.csv")
daten<- unname(read.csv2(datname, header=T,skip=0,dec="."))[,1]
library(weibullness)
fitres<- weibull.mle(daten)
fitres
fitres<- weibull.mle(daten,a=0.5)
fitres
fitres<- weibull.mle(daten,a=0.45)
fitres
str(fitres)
my_wb <- function(x, shape, scale, thres) {
dweibull(x - thres, shape, scale)
}
ml <- fitdistr(x, densfun = my_wb, start = list(
shape = pre_mle[1],
scale = pre_mle[2],
thres = pre_mle[3]
))
library(shiny)
library(survival)
library(survminer)
library(fitdistrplus)
library(WeibullR)
library(data.table)
library(ggplot2)
library(weibullness)
my_wb <- function(x, shape, scale, thres) {
dweibull(x - thres, shape, scale)
}
ml <- fitdistr(x, densfun = my_wb, start = list(
shape = pre_mle[1],
scale = pre_mle[2],
thres = pre_mle[3]
))
EPS <- sqrt(.Machine$double.eps)
llik_weibull <- function(shape, scale, thres, x) {
sum(dweibull(x - thres, shape, scale, log = TRUE))
}
thetahat_weibull <- function(x) {
if (any(x <= 0)) stop("x values must be positive")
toptim <- function(theta) -llik_weibull(theta[1], theta[2], theta[3], x)
mu <- mean(log(x))
sigma2 <- var(log(x))
shape.guess <- 1.2 / sqrt(sigma2)
scale.guess <- exp(mu + (0.572 / shape.guess))
thres.guess <- min(x) - 0.1 * sd(x)
res <- nlminb(c(shape.guess, scale.guess, thres.guess), toptim, lower = EPS)
c(shape = res$par[1], scale = res$par[2], thres = res$par[3])
}
llik_weibull <- function(shape, scale, thres, x) {
if (any(x <= thres)) return(-Inf)
sum(dweibull(x - thres, shape, scale, log = TRUE))
}
# MLE-Schätzung via nlminb()
thetahat_weibull <- function(x) {
EPS <- sqrt(.Machine$double.eps)
toptim <- function(theta) -llik_weibull(theta[1], theta[2], theta[3], x)
shape.guess <- 1.2 / sqrt(var(log(x)))
scale.guess <- exp(mean(log(x)) + 0.572 / shape.guess)
thres.guess <- min(x) - 0.1 * sd(x)
res <- nlminb(c(shape.guess, scale.guess, thres.guess), toptim, lower = c(EPS, EPS, -Inf))
names(res$par) <- c("shape", "scale", "threshold")
return(res$par)
}
x<- daten
# Schätzung ausführen
params <- thetahat_weibull(x)
params
library(MASS)
weibull3_density <- function(x, shape, scale, threshold) {
ifelse(x > threshold, dweibull(x - threshold, shape, scale), 0)
}
fit <- fitdistr(x, densfun = weibull3_density,
start = as.list(params),
lower = c(0.01, 0.01, min(x) - 1),
upper = c(10, 10, min(x) - 0.01))
daten
is.numeric(daten)
as.list(params)
llik_weibull3 <- function(shape, scale, threshold) {
if (any(x <= threshold)) return(-1e6)  # Strafe für ungültige Parameter
sum(dweibull(x - threshold, shape = shape, scale = scale, log = TRUE))
}
# MLE-Schätzung
fit <- mle2(llik_weibull3,
start = list(shape = 1.5, scale = 2, threshold = 0.1),
method = "Nelder-Mead")
library(bbmle)
llik_weibull3 <- function(shape, scale, threshold) {
if (any(x <= threshold)) return(-1e6)  # Strafe für ungültige Parameter
sum(dweibull(x - threshold, shape = shape, scale = scale, log = TRUE))
}
# MLE-Schätzung
fit <- mle2(llik_weibull3,
start = list(shape = 1.5, scale = 2, threshold = 0.1),
method = "Nelder-Mead")
# Ergebnisse
summary(fit)
confint(fit)
library(tpwb)
install.packages("tpwb")
library(tpwb)
tpwb(daten)
mlewb(daten)
mlewb(daten,2,3,1)
x<- rtpwb(1000,2,3,1) #n=1000 large sample
mlewb(x,2,3,1)
x<- rtpwb(1000,2,3,1) #n=1000 large sample
mlewb(x,20,30,1)
x<- rtpwb(1000,2,3,1) #n=1000 large sample
mlewb(x,7,5,10)
x<- rtpwb(1000,2,3,1) #n=1000 large sample
mlewb(x,4,5,2)
install.packages("ForestFit")
library(ForestFit)
fitWeibull(daten,TRUE,method="ml")
fitWeibull(daten,TRUE,method="ml",c(3,1000,0))
data("HW")
y<- data("HW")
y
y<- data(HW)
y
HW
data(DBH)
DBH
write.csv2(DBH,"o:/Eckhard/daten/multivar/DBH.csv")
DBH[55,]
EPS = sqrt(.Machine$double.eps) # "epsilon" for very small numbers
llik.weibull <- function(shape, scale, thres, x)
{
sum(dweibull(x - thres, shape, scale, log=T))
}
thetahat.weibull <- function(x)
{
if(any(x <= 0)) stop("x values must be positive")
toptim <- function(theta) -llik.weibull(theta[1], theta[2], theta[3], x)
mu = mean(log(x))
sigma2 = var(log(x))
shape.guess = 1.2 / sqrt(sigma2)
scale.guess = exp(mu + (0.572 / shape.guess))
thres.guess = 1
res = nlminb(c(shape.guess, scale.guess, thres.guess), toptim, lower=EPS)
c(shape=res$par[1], scale=res$par[2], thres=res$par[3])
}
pre_mle <- thetahat.weibull(daten)
pre_mle
my_wb <- function(x, shape, scale, thres) {
dweibull(x - thres, shape, scale)
}
ml <- fitdistr(daten, densfun = my_wb, start = list(shape = round(pre_mle[1], digits = 0), scale = round(pre_mle[2], digits = 0),
thres = round(pre_mle[3], digits = 0)))
ml
str(ml)
qnorm(0.025)
qchisq(0.975)
qnorm(0.975)
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
coef(fit_obj())
fit_obj()
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
mle()
mle
fitres
runApp('Eckhard/stat/R/zuverl/deskr_stat')
fit_obj()
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
fit_obj()
mle
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
fit_obj()$mle
ml
ci_left
ci_right
fit_obj
fit_obj()
pweibull3(201, 3, 1000, 200)
pweibull3(301, 3, 1000, 200)
pweibull3(1301, 3, 1000, 200)
runApp('Eckhard/stat/R/zuverl/deskr_stat')
sh
runApp('Eckhard/stat/R/zuverl/deskr_stat')
ml
sh
sc
th
runApp('Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('Eckhard/stat/R/zuverl/deskr_stat')
shiny::runApp('O:/Eckhard/stat/R/zuverl')
runApp('O:/Eckhard/stat/R/zuverl')
ausfall
zens
km_fit$time
km_fit$n.risk
km_fit$n.event
sum(km_fit$n.event)
sum(km_fit$n.censor)
runApp('Eckhard/stat/R/zuverl')
runApp('Eckhard/stat/R/zuverl')
runApp('Eckhard/stat/R/zuverl')
sprintf("%g",123)
sprintf("%g6",123)
sprintf("%.6g",123)
sprintf("%.6g",123456)
sprintf("%.6g",123456789)
sprintf("%.6g",0.123456789)
sprintf("%.6g",0.0000000000123456789)
runApp('Eckhard/stat/R/zuverl')
runApp('Eckhard/stat/R/zuverl')
runApp('Eckhard/stat/R/zuverl')
runApp('Eckhard/stat/R/zuverl')
runApp('Eckhard/stat/R/zuverl')
runApp('O:/Eckhard/stat/R/zuverl')
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
dl$time
dl$event
sum(dl$event)
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
cdat
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
install.packages("spatstat.univar")
runApp('Eckhard/stat/R/zuverl/zens_dstat')
datenr <- rweibull(100, shape = 2, scale = 5)
datenz<- as.numeric(runif(100)>=0.5)
df<- data.frame(zeit=datenr,id=datenz)
# Paket laden
library(survival)
library(survminer)
# Überlebensobjekt erstellen
surv_obj <- Surv(time = datenr, event = datenz)
# Kaplan-Meier-Schätzer berechnen
km_fit <- survfit(Surv(datenr, datenz) ~ 1, data=df)
View(km_fit)
km_fit$surv
km_fit$n.event[km_fit$n.event > 0] / km_fit$n.risk[km_fit$n.event > 0]
S <- c(0.95, 0.90, 0.85, 0.80)
c(1, S[-length(S)])
S[-1]
S[-2]
km_fit$n.event[km_fit$n.event > 0]*c(1,km_fit$surv[-50]) / km_fit$n.risk[km_fit$n.event > 0]
c(1,km_fit$surv[-50])
km_fit$surf-c(1,km_fit$surv[-100])
km_fit$surv-c(1,km_fit$surv[-100])
km_fit$surv[km_fit$n.event > 0]-c(1,km_fit$surv[km_fit$n.event > 0][-100])
km_fit$surv[km_fit$n.event > 0]-c(1,km_fit$surv[km_fit$n.event > 0][-sum(km_fit$n.event > 0)])
runApp('Eckhard/stat/R/zuverl/zens_dstat')
ff<- list(a=c(1,3,4,6),b=c(0,3,2,0))
ff$c<- c(2,1,2,1)
ff
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
dl$event
sum(dl$event)
runApp()
runApp()
sum(dl$event)
runApp('Eckhard/stat/R/zuverl/zens_dstat')
dl$time
e
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
ADt
ADt$Test
dl
ff
shiny::runApp('Eckhard/stat/R/zuverl/zens_dstat')
dl
dl$time
signif(ADt$Test[1],2)
ls()
print(dl)
test <- function(x) {
browser()
y <- x^2
return(y)
}
Q
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
datenr <- rweibull(100, shape = 2, scale = 5)
datenz<- as.numeric(runif(100)>=0.5)
df<- data.frame(zeit=datenr,id=datenz)
ADt<- ADcens(df$zeit,df$id,distr="weibull")
ADt
ADt$Test
ADt$Test[1]
unname(ADt$Test[1])
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
ADt
ff<- c(1,4,9,11,12,15)
match((ff>10))
which((ff>10))
which((ff>20))
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
runApp('Eckhard/stat/R/zuverl/zens_dstat')
shiny::runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
x
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
d
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp()
runApp()
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
gfun <- function(x) exp(D * (-x + v0))
print(gfun)
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
df1 <- data.frame(x = seq(-5, 5, length.out = 100), y = sin(seq(-5, 5, length.out = 100)), func = "sin(x)")
df2 <- data.frame(x = seq(-5, 5, length.out = 100), y = cos(seq(-5, 5, length.out = 100)), func = "cos(x)")
df3 <- data.frame(x = seq(-5, 5, length.out = 100), y = exp(seq(-5, 5, length.out = 100)), func = "exp(x)")
df_all <- bind_rows(df1, df2, df3)
library(ggplot2)
ggplot(df_all, aes(x = x, y = y, color = func)) +
geom_line() +
labs(color = "Funktion", x = "x", y = "y") +
theme_minimal()
load("~/Eckhard/forschung/abhaengigkeit/DA/R/worksp28102025.RData")
kendallst<- function(df,set1,set2,alpha=0.05){
#df..dataframe, alpha..Signifikanzniveau
#set1..Var.menge 1, dazu unabhaengige Variable werden aus Menge 2 gesucht
#Output: Indizes der unabhaengen Var.
k<- length(df[1,])  #Anzahl Variable
setr<- vector()
for (i in set2){
ff<- TRUE
for (j in set1){
ff<- ff &(cor.test(df[,i],df[,j],method="kendall")$p.value>=alpha)
}
if (ff) { setr<- c(setr,i)}
}
#return(km)
return(setr)
}
kendallst1<- function(df,i,j,alpha=0.05){
#df..dataframe, alpha..Signifikanzniveau
#i..Var. 1, j..Var. 2
#Output: Nichtsignifikanz = Unabhaengigkeit
ff<- (cor.test(df[,i],df[,j],method="kendall")$p.value>=alpha)
return(ff)
}
######korrigiertes Kendall tau
kendr0<- function(x,y){
k<- length(y)
ll<- kendr(x,y,out=1)  #Liste der Ergebnisse
l<- length(ll)
mx<- -1
for (l0 in 1:l){ #Schleife über Richtungen
d<- ll[[l0]]$dir
n0<- 0
for(i in 1:n){
for(j in 1:n){
if (i!=j){
if (all((d*x[i,])<=(d*x[j,]))){
n0<- 1
break
}}
}
if (n0==1) {break}
}
if (n0==1){  #normales Ergebnis
for (l in 1:length(ll)){
if (abs(ll[[l]]$dcoeff)>mx) {
mx<- abs(ll[[l]]$dcoeff)
}
}
}
}
return(mx)
}
list(time = c(1,4), v = c(3,9), cens = NULL)
shiny::runApp('Eckhard/stat/R/zuverl/beschleun_lebensd')
shiny::runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
runApp('O:/Eckhard/stat/R/zuverl/zens_dstat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
runApp('O:/Eckhard/stat/R/zuverl/beschleun_lebensd')
install.packages(ADGofTest)
install.packages("ADGofTest")
install.packages("e1071")
shiny::runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
shiny::runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
library(shiny); runApp('Eckhard/stat/R/zuverl/test2.R')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
runApp('O:/Eckhard/stat/R/zuverl/deskr_stat')
