Commit 6dcacac1 authored by matbuoro's avatar matbuoro
Browse files

Mise à jour 2016

parent 892c6eb0
......@@ -12,7 +12,7 @@ library(mcmcplots)
##-----------------------------INFO ----------------------------------##
year <- "2015"
year <- "2016"
site <- "Oir"
stade <- "smolt"
......@@ -22,6 +22,8 @@ work.dir<-paste("/media/ORE/Abundance",site,stade,sep="/")
setwd(work.dir)
##-----------------------------DATA ----------------------------------##
source(paste('data/data_',stade,'.R',sep="")) # creation du fichier Rdata
load(paste('data/data_',stade,"_",year,'.Rdata',sep="")) # chargement des données
......@@ -32,21 +34,29 @@ source(paste('parameters_',stade,'.R',sep="")) # chargement des paramètres
#------------------------INITS----------------------------------##
source(paste('inits/inits_',stade,'.R',sep="")) # création des inits des données
load(paste('inits/inits_',stade,'.Rdata',sep="")) # chargement des inits
if(site == "Bresle" && stade == "adult") {inits <- list(read.bugsdata(paste("inits/init-",site,"-",stade,year,".txt",sep="")))}
if(site == "Nivelle") {inits <- list(read.bugsdata(paste("inits/init-",site,"-",stade,year,".txt",sep="")))}
#if(!file.exists(paste('inits/inits_',stade,year,'.Rdata',sep=""))){
if(!file.exists(paste("inits/init-",site,"-",stade,year,".txt",sep=""))){
source(paste('inits/inits_',stade,'.R',sep="")) # création des inits des données
#load(paste('inits/inits_',stade,year,'.Rdata',sep=""))
}
#load(paste('inits/inits_',stade,'.Rdata',sep="")) # chargement des inits
#if(site == "Bresle" && stade == "adult") {inits <- list(read.bugsdata(paste("inits/init-",site,"-",stade,year,".txt",sep="")))}
#if(site == "Nivelle") {inits <- list(read.bugsdata(paste("inits/init-",site,"-",stade,year,".txt",sep="")))}
inits <- list(read.bugsdata(paste("inits/init-",site,"-",stade,year,".txt",sep="")))
#------------------------MODEL----------------------------------##
model <- paste("model/",stade,"-",site,".R",sep="") # path of the model
model <- paste("model/model_",stade,"-",site,".R",sep="") # path of the model
if(site == "Scorff" && stade == "smolt") {model <- paste("model/",stade,"-",site,"_",year,".R",sep="")} # le modèle Scorrf pour les smolt peut changer tous les ans suivant conditions
model
filename <- file.path(work.dir, model)
#system(paste("cp",model,paste(stade,"-",site,".txt",sep=""),sep=""))
#---------------------------ANALYSIS-----------------------------##
nChains = length(inits) # Number of chains to run.
adaptSteps = 1000 # Number of steps to "tune" the samplers.
nburnin=5000 # Number of steps to "burn-in" the samplers.
nstore=50000 # Total number of steps in chains to save.
nburnin=500 # Number of steps to "burn-in" the samplers.
nstore=1000 # Total number of steps in chains to save.
nthin=1 # Number of steps to "thin" (1=keep every step).
#nPerChain = ceiling( ( numSavedSteps * thinSteps ) / nChains ) # Steps per chain.
......@@ -57,15 +67,23 @@ start.time = Sys.time(); cat("Start of the run\n");
fit <- bugs(
data
,inits
,model.file = model
,model.file = filename
,parameters
,n.chains = nChains, n.iter = nstore + nburnin, n.burnin = nburnin, n.thin = nthin
,DIC=FALSE
,codaPkg = FALSE, clearWD=TRUE
,codaPkg = FALSE, clearWD=FALSE
#,debug=TRUE
,working.directory=work.dir
,working.directory=paste(work.dir,"bugs",sep="/")
)
## cleaning
system("rm bugs/CODA*")
# save last values for inits
#inits <- fit$last.values
#if(site == "Nivelle") {save(inits,file=paste('inits/inits_',stade,year,'.Rdata',sep=""))}
#bugs.inits(inits, n.chains=1,digits=3, inits.files = paste('inits/init-',site,'-',stade,year,'.txt',sep=""))
######### JAGS ##########
## Compile & adapt
......
list(Nyears=3.10000E+01, C_MC=c(5.08000E+02, 2.75000E+02, 2.96000E+02, 5.41000E+02, 7.42000E+02, 1.50000E+02, 5.74000E+02, 2.08000E+02, 3.27000E+02, 6.18000E+02, 7.64000E+02, 2.02000E+02, 5.20000E+02, 1.95000E+02, 1.83700E+03, 6.81000E+02, 1.86800E+03, 8.40000E+02, 8.48000E+02, 6.58000E+02, 8.87000E+02, 7.27000E+02, 1.24200E+03, 1.82800E+03, 6.86000E+02, 9.99000E+02, 9.72000E+02, 5.72000E+02, 7.39000E+02, 1.24800E+03, 5.08000E+02), Cm_MC=c(8.90000E+01, 3.10000E+01, 5.90000E+01, 6.50000E+01, 3.80000E+01, 3.50000E+01, 5.00000E+01, 2.60000E+01, 1.70000E+01, 6.30000E+01, 7.60000E+01, 6.80000E+01, 9.20000E+01, 5.90000E+01, 2.87000E+02, 2.27000E+02, 4.42000E+02, 3.26000E+02, 2.69000E+02, 2.52000E+02, 3.40000E+02, 1.83000E+02, 3.31000E+02, 3.44000E+02, 1.69000E+02, 3.77000E+02, 1.87000E+02, 2.68000E+02, 2.64000E+02, 3.60000E+02, 1.89000E+02), Cm_R=c(6.10000E+01, 2.40000E+01, 4.30000E+01, 4.30000E+01, 3.50000E+01, 2.70000E+01, 4.30000E+01, 2.40000E+01, 1.00000E+01, 5.30000E+01, 5.80000E+01, 2.70000E+01, 4.40000E+01, 4.50000E+01, 2.46000E+02, 1.13000E+02, 3.52000E+02, 2.22000E+02, 1.86000E+02, 1.98000E+02, 2.23000E+02, 1.41000E+02, 1.89000E+02, 1.82000E+02, 8.10000E+01, 2.07000E+02, 1.34000E+02, 1.17000E+02, 1.22000E+02, 1.93000E+02, 4.40000E+01), D_MC=c(1.00000E+00, 6.00000E+00, 0.00000E+00, 2.00000E+00, 6.00000E+00, 0.00000E+00, 0.00000E+00, 0.00000E+00, 1.30000E+01, 1.00000E+00, 7.00000E+00, 5.00000E+00, 1.00000E+00, 2.00000E+00, 4.00000E+00, 0.00000E+00, 1.00000E+00, 9.00000E+00, 1.00000E+00, 1.00000E+00, 0.00000E+00, 1.00000E+00, 4.00000E+00, 0.00000E+00, 0.00000E+00, 1.00000E+00, 0.00000E+00, 0.00000E+00, 0.00000E+00, 0.00000E+00, 1.00000E+00), eff_MC=c(1.00000E+00, 1.00000E+00, 3.00000E+00, 4.00000E+00, 2.00000E+00, 2.00000E+00, 2.00000E+00, 1.00000E+00, 2.00000E+00, 3.00000E+00, 3.00000E+00, 1.20000E+01, 5.00000E+00, 6.00000E+00, 1.20000E+01, 1.20000E+01, 2.10000E+01, 2.20000E+01, 2.10000E+01, 1.30000E+01, 1.80000E+01, 1.10000E+01, 1.70000E+01, 1.60000E+01, 7.00000E+00, 1.10000E+01, 8.00000E+00, 1.20000E+01, 1.30000E+01, 1.80000E+01, 1.10000E+01), Q_MC=c(1.23678E+03, 1.03033E+03, 1.13633E+03, 1.12410E+03, 5.68200E+02, 6.63533E+02, 6.43367E+02, 6.95067E+02, 1.77667E+03, 1.19507E+03, 5.15133E+02, 4.71033E+02, 1.01723E+03, 1.39567E+03, 1.97700E+03, 3.80900E+03, 1.12233E+03, 1.04057E+03, 6.20233E+02, 8.98800E+02, 1.22780E+03, 1.19723E+03, 1.43733E+03, 8.41033E+02, 8.21172E+02, 5.06143E+02, 7.97600E+02, 1.59720E+03, 1.20100E+03, 1.48000E+03, 2.00640E+03))
list(log_cess_MC=-2.22400E-01, rate_lambda=2.48000E-03, shape_lambda=2.89300E+00, logit_flow_MC=-5.50000E-01, logit_int_MC=7.00000E-01, Ntot=c(3.48000E+02, 2.12000E+02, 2.15000E+02, 3.57000E+02, 6.83000E+02, 1.15000E+02, 4.93000E+02, 1.92000E+02, 1.92000E+02, 5.19000E+02, 5.83000E+02, 8.00000E+01, 2.48000E+02, 1.48000E+02, 1.57400E+03, 3.39000E+02, 1.48700E+03, 5.72000E+02, 5.86000E+02, 5.17000E+02, 5.81000E+02, 5.60000E+02, 7.09000E+02, 9.67000E+02, 3.28000E+02, 5.48000E+02, 6.96000E+02, 2.49000E+02, 3.41000E+02, 6.69000E+02, 1.18000E+02), lambda=c(3.48000E+02, 2.12000E+02, 2.15000E+02, 3.57000E+02, 6.83000E+02, 1.15000E+02, 4.93000E+02, 1.92000E+02, 1.92000E+02, 5.19000E+02, 5.83000E+02, 8.00000E+01, 2.48000E+02, 1.48000E+02, 1.57400E+03, 3.39000E+02, 1.48700E+03, 5.72000E+02, 5.86000E+02, 5.17000E+02, 5.81000E+02, 5.60000E+02, 7.09000E+02, 9.67000E+02, 3.28000E+02, 5.48000E+02, 6.96000E+02, 2.49000E+02, 3.41000E+02, 6.69000E+02, 1.18000E+02), p_MC=c(6.85400E-01, 7.74200E-01, 7.28800E-01, 6.61500E-01, 9.21100E-01, 7.71400E-01, 8.60000E-01, 9.23100E-01, 5.88200E-01, 8.41300E-01, 7.63200E-01, 3.97100E-01, 4.78300E-01, 7.62700E-01, 8.57100E-01, 4.97800E-01, 7.96400E-01, 6.81000E-01, 6.91400E-01, 7.85700E-01, 6.55900E-01, 7.70500E-01, 5.71000E-01, 5.29100E-01, 4.79300E-01, 5.49100E-01, 7.16600E-01, 4.36600E-01, 4.62100E-01, 5.36100E-01, 2.32800E-01))
OpenBUGS version 3.2.3 rev 1012
model is syntactically correct
data loaded
model compiled
initial values loaded but chain contain uninitialized variables
initial values generated, model initialized
500 updates took 0 s
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
monitor set
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
inference can not be made when sampler is in adaptive phase
1000 updates took 1 s
CODA files written
no monitors set
modelCheck('/home/basp-meco88/Documents/RESEARCH/PROJECTS/ORE/Abundance/Oir/smolt/bugs/model_smolt-Oir.R.txt')
modelData('/home/basp-meco88/Documents/RESEARCH/PROJECTS/ORE/Abundance/Oir/smolt/bugs/data.txt')
modelCompile(1)
modelSetRN(1)
modelInits('/home/basp-meco88/Documents/RESEARCH/PROJECTS/ORE/Abundance/Oir/smolt/bugs/inits1.txt',1)
modelGenInits()
modelUpdate(500,1,500)
samplesSet(logit_int_MC)
samplesSet(logit_flow_MC)
samplesSet(log_cess_MC)
samplesSet(shape_lambda)
samplesSet(rate_lambda)
samplesSet(mean_gamma)
samplesSet(var_gamma)
samplesSet(lambda)
samplesSet(Ntot)
samplesSet(Nesc)
samplesSet(overdisp_MC)
samplesSet(mean_MC)
samplesSet(p_MC)
samplesSet(alpha_MC)
samplesSet(beta_MC)
samplesSet(test)
summarySet(logit_int_MC)
summarySet(logit_flow_MC)
summarySet(log_cess_MC)
summarySet(shape_lambda)
summarySet(rate_lambda)
summarySet(mean_gamma)
summarySet(var_gamma)
summarySet(lambda)
summarySet(Ntot)
summarySet(Nesc)
summarySet(overdisp_MC)
summarySet(mean_MC)
summarySet(p_MC)
summarySet(alpha_MC)
summarySet(beta_MC)
summarySet(test)
modelUpdate(1000,1,1000)
samplesCoda('*', '/home/basp-meco88/Documents/RESEARCH/PROJECTS/ORE/Abundance/Oir/smolt/bugs//')
summaryStats('*')
modelQuit('y')
list(
## METTRE A JOUR
Ntot = c(
792.0,365.0,433.0,780.0,886.0,
237.0,698.0,237.0,468.0,770.0,
1133.0,482.0,1111.0,255.0,2059.0,
1413.0,2350.0,1242.0,1159.0,809.0,
1321.0,962.0,2228.0,3502.0,1544.0,
1952.0,1358.0,1358.0,1647.0,2500,
2000),
## METTRE A JOUR
lambda = c(
762.7,361.5,488.6,749.6,885.7,
231.4,723.4,225.5,440.7,761.1,
1089.0,489.0,1092.0,277.8,2064.0,
1448.0,2313.0,1284.0,1145.0,820.3,
1371.0,962.5,2208.0,3465.0,1588.0,
1955.0,1409.0,1369.0,1650.0,2500,
2000)),
## METTRE A JOUR
p_MC = c(
0.6704,0.8023,0.6877,0.6906,0.8445,
0.6361,0.8376,0.8768,0.6882,0.818,
0.6731,0.4599,0.4611,0.7612,0.8879,
0.4909,0.8072,0.6636,0.7157,0.816,
0.6614,0.7451,0.5876,0.5282,0.4533,
0.4882,0.7201,0.4379,0.4570,0.45,
0.25),
## NO UPDATE
log_cess_MC = -0.2224,
rate_lambda = 0.00248,
shape_lambda = 2.893,
logit_flow_MC = -0.55,
logit_int_MC = 0.7
)
list(log_cess_MC=-2.224E-01, rate_lambda=2.480E-03, shape_lambda=2.893E+00, logit_flow_MC=-5.500E-01, logit_int_MC=7.000E-01, Ntot=c(3.480E+02, 2.120E+02, 2.150E+02, 3.570E+02, 6.830E+02, 1.150E+02, 4.930E+02, 1.920E+02, 1.920E+02, 5.190E+02, 5.830E+02, 8.000E+01, 2.480E+02, 1.480E+02, 1.574E+03, 3.390E+02, 1.487E+03, 5.720E+02, 5.860E+02, 5.170E+02, 5.810E+02, 5.600E+02, 7.090E+02, 9.670E+02, 3.280E+02, 5.480E+02, 6.960E+02, 2.490E+02, 3.410E+02, 6.690E+02, 1.180E+02), lambda=c(3.480E+02, 2.120E+02, 2.150E+02, 3.570E+02, 6.830E+02, 1.150E+02, 4.930E+02, 1.920E+02, 1.920E+02, 5.190E+02, 5.830E+02, 8.000E+01, 2.480E+02, 1.480E+02, 1.574E+03, 3.390E+02, 1.487E+03, 5.720E+02, 5.860E+02, 5.170E+02, 5.810E+02, 5.600E+02, 7.090E+02, 9.670E+02, 3.280E+02, 5.480E+02, 6.960E+02, 2.490E+02, 3.410E+02, 6.690E+02, 1.180E+02), p_MC=c(6.854E-01, 7.742E-01, 7.288E-01, 6.615E-01, 9.211E-01, 7.714E-01, 8.600E-01, 9.231E-01, 5.882E-01, 8.413E-01, 7.632E-01, 3.971E-01, 4.783E-01, 7.627E-01, 8.571E-01, 4.978E-01, 7.964E-01, 6.810E-01, 6.914E-01, 7.857E-01, 6.559E-01, 7.705E-01, 5.710E-01, 5.291E-01, 4.793E-01, 5.491E-01, 7.166E-01, 4.366E-01, 4.621E-01, 5.361E-01, 2.328E-01))
......@@ -25,47 +25,52 @@ inits_fix <- list(
###################################################
# TO UPDATE
###################################################
inits_updated <- list(
#inits_updated <- list(
## METTRE A JOUR
# p_MC = c(
# 0.6704,0.8023,0.6877,0.6906,0.8445,
# 0.6361,0.8376,0.8768,0.6882,0.818,
# 0.6731,0.4599,0.4611,0.7612,0.8879,
# 0.4909,0.8072,0.6636,0.7157,0.816,
# 0.6614,0.7451,0.5876,0.5282,0.4533,
# 0.4882,0.7201,0.4379,0.4570,0.45,
# 0.25)
# )
p_MC = (data$Cm_R / data$Cm_MC)
## METTRE A JOUR
Ntot = c(
792.0,365.0,433.0,780.0,886.0,
237.0,698.0,237.0,468.0,770.0,
1133.0,482.0,1111.0,255.0,2059.0,
1413.0,2350.0,1242.0,1159.0,809.0,
1321.0,962.0,2228.0,3502.0,1544.0,
1952.0,1358.0,1358.0,1647.0,2500,
2000),
# Ntot = c(
# 792.0,365.0,433.0,780.0,886.0,
# 237.0,698.0,237.0,468.0,770.0,
# 1133.0,482.0,1111.0,255.0,2059.0,
# 1413.0,2350.0,1242.0,1159.0,809.0,
# 1321.0,962.0,2228.0,3502.0,1544.0,
# 1952.0,1358.0,1358.0,1647.0,2500,
# 2000),
Ntot = as.integer(data$C_MC * p_MC)
## METTRE A JOUR
lambda = c(
762.7,361.5,488.6,749.6,885.7,
231.4,723.4,225.5,440.7,761.1,
1089.0,489.0,1092.0,277.8,2064.0,
1448.0,2313.0,1284.0,1145.0,820.3,
1371.0,962.5,2208.0,3465.0,1588.0,
1955.0,1409.0,1369.0,1650.0,2500,
2000),
# lambda = c(
# 762.7,361.5,488.6,749.6,885.7,
# 231.4,723.4,225.5,440.7,761.1,
# 1089.0,489.0,1092.0,277.8,2064.0,
# 1448.0,2313.0,1284.0,1145.0,820.3,
# 1371.0,962.5,2208.0,3465.0,1588.0,
# 1955.0,1409.0,1369.0,1650.0,2500,
# 2000),
lambda = Ntot
## METTRE A JOUR
p_MC = c(
0.6704,0.8023,0.6877,0.6906,0.8445,
0.6361,0.8376,0.8768,0.6882,0.818,
0.6731,0.4599,0.4611,0.7612,0.8879,
0.4909,0.8072,0.6636,0.7157,0.816,
0.6614,0.7451,0.5876,0.5282,0.4533,
0.4882,0.7201,0.4379,0.4570,0.45,
0.25)
inits_updated <- list(
Ntot = Ntot
, lambda = lambda
, p_MC = p_MC
)
# inits_updated <- list(
# Ntot = Ntot
# , lambda = lambda
# , logit_pi = logit_pi
# )
inits <- list(c( inits_fix,inits_updated))
save(inits,file=paste(paste('inits/inits_',stade,'.Rdata',sep="")))
#save(inits,file=paste('inits/inits_',stade,year,'.Rdata',sep=""))
bugs.inits(inits, n.chains=1,digits=3, inits.files = paste('inits/init-',site,'-',stade,year,'.txt',sep=""))
=============================
DIAGNOSTICS
=============================
---------------------------
Heidelberger and Welch's convergence diagnostic
heidel.diag is a run length control diagnostic based on a criterion of relative accuracy for the estimate of the mean. The default setting corresponds to a relative accuracy of two significant digits.
heidel.diag also implements a convergence diagnostic, and removes up to half the chain in order to ensure that the means are estimated from a chain that has converged.
Stationarity start p-value
test iteration
logit_int_MC passed 1 0.7184
logit_flow_MC passed 1 0.0526
log_cess_MC passed 1 0.2295
shape_lambda passed 1 0.8390
rate_lambda passed 1 0.8781
mean_gamma passed 1 0.7485
var_gamma passed 1 0.6012
Halfwidth Mean Halfwidth
test
logit_int_MC passed 5.07e-01 1.00e-02
logit_flow_MC failed -9.75e-02 1.14e-02
log_cess_MC failed 1.03e-01 2.82e-02
shape_lambda passed 2.48e+00 1.02e-01
rate_lambda passed 2.08e-03 9.31e-05
mean_gamma passed 1.21e+03 8.99e+00
var_gamma passed 6.28e+05 3.46e+04
---------------------------
Geweke's convergence diagnostic
Geweke (1992) proposed a convergence diagnostic for Markov chains based on a test for equality of the means of the first and last part of a Markov chain (by default the first 10% and the last 50%).
If the samples are drawn from the stationary distribution of the chain, the two means are equal and Geweke's statistic has an asymptotically standard normal distribution.
The test statistic is a standard Z-score: the difference between the two sample means divided by its estimated standard error. The standard error is estimated from the spectral density at zero and so takes into account any autocorrelation.
The Z-score is calculated under the assumption that the two parts of the chain are asymptotically independent, which requires that the sum of frac1 and frac2 be strictly less than 1.
Fraction in 1st window = 0.1
Fraction in 2nd window = 0.5
logit_int_MC logit_flow_MC log_cess_MC shape_lambda rate_lambda mean_gamma var_gamma
0.262 1.687 -3.009 0.433 0.419 -0.502 -0.640
---------------------------
Raftery and Lewis's diagnostic
Quantile (q) = 0.025
Accuracy (r) = +/- 0.005
Probability (s) = 0.95
You need a sample size of at least 3746 with these values of q, r and s
=============================
DIAGNOSTICS
=============================
---------------------------
Heidelberger and Welch's convergence diagnostic
heidel.diag is a run length control diagnostic based on a criterion of relative accuracy for the estimate of the mean. The default setting corresponds to a relative accuracy of two significant digits.
heidel.diag also implements a convergence diagnostic, and removes up to half the chain in order to ensure that the means are estimated from a chain that has converged.
Stationarity start p-value
test iteration
sd_prem failed NA 2.77e-06
mup_rem passed 1 3.11e-01
int_width passed 301 4.10e-01
width_coef passed 301 5.29e-01
rate_lcpu passed 301 3.75e-01
sigma_dOir passed 1 2.68e-01
sigma_yOir passed 1 9.52e-01
sigma_gryrOir passed 1 3.67e-01
coef_PC passed 1 8.23e-01
Halfwidth Mean Halfwidth
test
sd_prem <NA> NA NA
mup_rem passed 0.805 0.00244
int_width passed -0.764 0.03540
width_coef passed 0.791 0.03054
rate_lcpu passed 0.560 0.04292
sigma_dOir passed 0.718 0.00326
sigma_yOir passed 1.272 0.03306
sigma_gryrOir passed 0.912 0.01078
coef_PC passed 0.134 0.00874
---------------------------
Geweke's convergence diagnostic
Geweke (1992) proposed a convergence diagnostic for Markov chains based on a test for equality of the means of the first and last part of a Markov chain (by default the first 10% and the last 50%).
If the samples are drawn from the stationary distribution of the chain, the two means are equal and Geweke's statistic has an asymptotically standard normal distribution.
The test statistic is a standard Z-score: the difference between the two sample means divided by its estimated standard error. The standard error is estimated from the spectral density at zero and so takes into account any autocorrelation.
The Z-score is calculated under the assumption that the two parts of the chain are asymptotically independent, which requires that the sum of frac1 and frac2 be strictly less than 1.
Fraction in 1st window = 0.1
Fraction in 2nd window = 0.5
sd_prem mup_rem int_width width_coef rate_lcpu sigma_dOir sigma_yOir sigma_gryrOir coef_PC
5.2540 1.0550 -1.7698 2.3750 0.0962 1.0448 0.5791 -0.4697 0.4144
---------------------------
Raftery and Lewis's diagnostic
Quantile (q) = 0.025
Accuracy (r) = +/- 0.005
Probability (s) = 0.95
You need a sample size of at least 3746 with these values of q, r and s
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment