Skip to main content

Advertisement

Table 8 (Ex-Sect-9.4.1). Mean absolute error (MAE) in the estimation of four statistics (first component) and normalized computing time (Continued)

From: Adaptive independent sticky MCMC algorithms

Technique T N G Init. MAE     Avg. MAE Time
     Mean Variance Skewness Kurtosis   
MH (σ p =10) 1 10000 In2 0.178 0.126 0.091 0.012 0.102 0.162
   20000   0.151 0.112 0.090 0.008 0.090 0.331
   30000   0.138 0.063 0.068 0.007 0.069 0.492
  2 10000   0.130 0.062 0.066 0.006 0.066 0.196
  3    0.125 0.066 0.063 0.006 0.065 0.223
  10 2000   0.149 0.083 0.075 0.009 0.079 0.081
Adaptive MH 10 2000 In2 0.158 0.082 0.087 0.012 0.084 0.090
  100    0.146 0.076 0.073 0.010 0.076 0.634
HMC 10 2000 In2 0.152 0.092 0.079 0.015 0.084 0.092
  100    0.148 0.081 0.070 0.012 0.077 0.630
Slice 3 2000 In2 0.204 0.105 0.103 0.022 0.108 0.156
  10    0.188 0.091 0.095 0.018 0.098 0.463
  3 10000   0.132 0.051 0.066 0.007 0.064 0.783
  1. All the techniques are used within a Gibbs sampler: N G is the number of iterations of the Gibbs sampler whereas T is is the number of iterations of the technique within Gibbs (so that T×N G is the global number of MCMC iterations). The best results (in each column, and in each panel) are highlighted with italics