Module 12:Application of stochastic processes in areas of engineering and management science
Lecture 40:Use of Markov Chain Monte Carlo Method
Results using MCMC Using MCMC we obtain the average values of , variance and bias for the GE distribution. We also report the bootstrap values of the confidence intervals. For calculating the average values of we consider a sample size n of 200 and conduct 200 runs for all the models. Thus in Table 12.1, 12.2 we report our simulation runs results using MCMC method.
Table 12.1: Average values, variance and bias of
GED complete data without EM and using MCMC
=1
=2
=1.5
Average
1.009
2.001
1.501
Variance
0.009
0.027
0.016
Bias
0.073
0.134
0.097
GED incomplete data without EM and using MCMC
=1
=2
=1.5
Average
1.010
1.999
1.521
Variance
0.008
0.023
0.009
Bias
0.074
0.126
0.080
Table 12.2: Bootstrap-1 and bootstrap-1 confidence intervals for , and for both complete and masked data using EM algorithm and without EM algorithm and using MCMC
GED incomplete data without using EM and using MCMC
, variance and bias for the GE distribution. We also report the bootstrap values of the confidence intervals. For calculating the average values of 
we consider a sample size n of 200 and conduct 200 runs for all the models. Thus in Table 12.1, 12.2 we report our simulation runs results using MCMC method.
=1
=2
=1.5
=1
=2
=1.5
, 
and 
for both complete and masked data using EM algorithm and without EM algorithm and using MCMC
