paul-corbalan
/
Scan-Statistics-Project-4Y-INSA
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fusion of codes

This commit is contained in:
Paul-Corbalan 2022-05-21 18:15:56 +02:00
parent 07f1cac62a
commit 68cbc5498d
1 changed files with 9 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
Comparaison_of_methods.rmd
View File

@ -354,30 +354,27 @@ plot_graph_distrib_score(distrib_score_theo, distrib_score_mc)
LocalScoreMC <- function(lambda0, lambda1, NbSeq, T, X_seq, P_X, tbe0){ LocalScoreMC <- function(lambda0, lambda1, NbSeq, T, X_seq, P_X, tbe0){
E = ComputeE(lambda0, lambda1) E = ComputeE(lambda0, lambda1)
pvalue = c() pvalue = c()
X = c()
min_X = min(X_seq) min_X = min(X_seq)
max_X = max(X_seq) max_X = max(X_seq)
NbSeq.NonNulles = 0 NbSeq.NonNulles = 0
for (i in 1:NbSeq){ for (i in 1:NbSeq){
x = floor(E*log(dexp(tbe0[[i]], rate = lambda1)/dexp(tbe0[[i]], rate = lambda0))) x = floor(E*log(dexp(tbe0[[i]], rate = lambda1)/dexp(tbe0[[i]], rate = lambda0)))
if (length(x)!=0){ if (length(x)!=0){
X = c(X,x)
LS = localScoreC(x)$localScore[1] LS = localScoreC(x)$localScore[1]
daudin_result = daudin(localScore = LS, score_probabilities = P_X, sequence_length = length(x), sequence_min = min_X, sequence_max = max_X) daudin_result = daudin(localScore = LS, score_probabilities = P_X, sequence_length = length(x), sequence_min = min_X, sequence_max = max_X)
options(warn = -1) # Disable warnings print options(warn = -1) # Disable warnings print
pvalue = c(pvalue, daudin_result) pvalue = c(pvalue, daudin_result)
NbSeq.NonNulles = NbSeq.NonNulles + 1 NbSeq.NonNulles = NbSeq.NonNulles + 1
} }
} }
LS_H0=data.frame(num=1:NbSeq.NonNulles, pvalue_scan=pvalue, class=(pvalue<0.05)*1) LS_H0=data.frame(num=1:NbSeq.NonNulles, pvalue_scan=pvalue, class=as.numeric(pvalue<0.05))
return(LS_H0) return(LS_H0)
} }
``` ```
## 4. Experience plan for comparaison ## 4. Experience plan for comparaison
```{r} ```{r}
CompareMethods <- function(lambda0, lambda1, NbSeq, T, tau){ CompareMethods <- function(lambda0, lambda1, NbSeq, T, tau_scan, tau_H1){
if (lambda0 < lambda1){ if (lambda0 < lambda1){
cat("For T = ", T, ", Nb = ", NbSeq, ", lambda0 = ", lambda0, " and lambda1 = ", lambda1, ":\n", sep = "") cat("For T = ", T, ", Nb = ", NbSeq, ", lambda0 = ", lambda0, " and lambda1 = ", lambda1, ":\n", sep = "")
@ -395,7 +392,7 @@ CompareMethods <- function(lambda0, lambda1, NbSeq, T, tau){
tbe0[[i]] = tbei tbe0[[i]] = tbei
#Simulation for sequences under H1 #Simulation for sequences under H1
ppj1 = SimulationH1(lambda0, lambda1, T, 3) ppj1 = SimulationH1(lambda0, lambda1, T, tau_H1)
nj = length(ppj1) nj = length(ppj1)
pp1[[i]] = ppj1 pp1[[i]] = ppj1
tbej = ppj1[2:nj]-ppj1[1:nj-1] tbej = ppj1[2:nj]-ppj1[1:nj-1]
@ -409,9 +406,9 @@ CompareMethods <- function(lambda0, lambda1, NbSeq, T, tau){
LS_obtained = c(LS_H0$class, LS_H1$class) LS_obtained = c(LS_H0$class, LS_H1$class)
options(warn = -1) options(warn = -1)
Emp = EmpDistrib(lambda0,10**5,T,tau) Emp = EmpDistrib(lambda0, 10**5, T, tau_scan)
SS_H0 = ScanStatMC(NbSeq, T, tau, Emp, pp0) SS_H0 = ScanStatMC(NbSeq, T, tau_scan, Emp, pp0)
SS_H1 = ScanStatMC(NbSeq, T, tau, Emp, pp1) SS_H1 = ScanStatMC(NbSeq, T, tau_scan, Emp, pp1)
SS_obtained = c(SS_H0$class, SS_H1$class) SS_obtained = c(SS_H0$class, SS_H1$class)
@ -450,7 +447,8 @@ CompareMethods <- function(lambda0, lambda1, NbSeq, T, tau){
```{r} ```{r}
NbSeq = 10**4 NbSeq = 10**4
T = 10 T = 10
tau = 2 tau_scan = 2
tau_H1 = 3
list_of_lambda = list() list_of_lambda = list()
list_of_lambda[[1]] = c(1, 3) list_of_lambda[[1]] = c(1, 3)
@ -465,7 +463,7 @@ legend_list = c()
for (Lambda in list_of_lambda){ for (Lambda in list_of_lambda){
lambda0 = Lambda[1] lambda0 = Lambda[1]
lambda1 = Lambda[2] lambda1 = Lambda[2]
result = CompareMethods(lambda0, lambda1, NbSeq, T, tau) result = CompareMethods(lambda0, lambda1, NbSeq, T, tau_scan)
title_ROC = TeX(paste(r'(ROC curve for several values of $\lambda_0$ and $\lambda_1$)')) title_ROC = TeX(paste(r'(ROC curve for several values of $\lambda_0$ and $\lambda_1$)'))
perfSS = result[1] perfSS = result[1]