Commit 4dd60346 authored by matbuoro's avatar matbuoro
Browse files

Mise à jour summary

parent 72917d98
=============================
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
sigmapi_MP passed 1 0.29851
sigmapi_R failed NA 0.00127
mupi_oF passed 1 0.14574
sigmapi_oF passed 1 0.28920
diffF_1SW passed 1 0.11100
diffF_MSW passed 1 0.11554
diff1SW passed 501 0.05261
diffMSW passed 1 0.40440
pi_oD passed 1 0.13998
shape_lambda passed 1 0.89298
rate_lambda passed 1 0.89803
lambda_tot0 passed 1 0.31651
Halfwidth Mean Halfwidth
test
sigmapi_MP passed 0.3491 2.25e-02
sigmapi_R <NA> NA NA
mupi_oF passed 0.7872 3.75e-02
sigmapi_oF failed 1.1734 5.64e-01
diffF_1SW failed -0.1038 1.53e-02
diffF_MSW passed -0.8611 2.92e-02
diff1SW failed 0.5074 1.06e-01
diffMSW failed 0.9013 1.11e-01
pi_oD passed 0.2146 1.72e-02
shape_lambda passed 6.4713 3.11e-01
rate_lambda passed 0.0104 5.17e-04
lambda_tot0 passed 616.8054 2.77e+01
---------------------------
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
sigmapi_MP sigmapi_R mupi_oF sigmapi_oF diffF_1SW diffF_MSW diff1SW diffMSW pi_oD
-0.612 0.448 2.261 0.633 2.681 3.960 3.518 0.110 -2.038
shape_lambda rate_lambda lambda_tot0
0.316 0.215 -1.870
---------------------------
Raftery and Lewis's diagnostic
Quantile (q) = 0.025
Accuracy (r) = +/- 0.005
Probability (s) = 0.95
Burn-in Total Lower bound Dependence
(M) (N) (Nmin) factor (I)
sigmapi_MP 108 101576 3746 27.10
sigmapi_R 12 17367 3746 4.64
mupi_oF 12 13347 3746 3.56
sigmapi_oF 91 101052 3746 27.00
diffF_1SW 8 8928 3746 2.38
diffF_MSW 8 9864 3746 2.63
diff1SW 20 21575 3746 5.76
diffMSW 25 26975 3746 7.20
pi_oD 24 22008 3746 5.88
shape_lambda 15 16561 3746 4.42
rate_lambda 22 23208 3746 6.20
lambda_tot0 16 18154 3746 4.85
e_1SW <- as.matrix(fit.mcmc[,paste("e_1SW[",1:22,"]",sep="")])
e_MSW <- as.matrix(fit.mcmc[,paste("e_MSW[",1:22,"]",sep="")])
e_1SW <- as.matrix(fit.mcmc[,paste("e_1SW[",1:data$Y,"]",sep="")])
e_MSW <- as.matrix(fit.mcmc[,paste("e_MSW[",1:data$Y,"]",sep="")])
write.csv(e_1SW,file="results/e_1SW.csv",sep=";")
write.csv(e_MSW,file="results/e_MSW.csv",sep=";")
\ No newline at end of file
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