Table 4 2ln(Bayes Factors) and posterior probabilities of each model considered for the three data sets For a given data set, the first five rows of data correspond to the model under H1 while the first five columns correspond to the model under H0. The (i, j)th element in the matrix represents the value of 2ln(Bayes Factors) for the model corresponding to the i th row versus the model corresponding to the j th column. Values of 2ln(Bayes Factors) are in bold print if they exceed 2. The last column provides values of the posterior probability of the model in the i th row. Those values corresponding to selected models are in bold print.
From: A new mixture model approach to analyzing allelic-loss data using Bayes factors
Barrett data set | ||||||
|---|---|---|---|---|---|---|
H1/H0 | 2 bb* | 2 bb/bin | 2 bin | 1 bb | 1 bin | Post.Prob** |
2 bb | 0 | -4.398 | -0.114 | 12.144 | 45.281 | 0.090 |
2 bb/bin | 4.398 | 0 | 4.284 | 16.542 | 49.679 | 0.814 |
2 bin | 0.114 | -4.284 | 0 | 12.258 | 45.395 | 0.096 |
1 bb | -12.144 | -16.542 | -12.258 | 0 | 33.137 | < 0.001 |
1 bin | -45.281 | -49.679 | -45.395 | -33.137 | 0 | < 0.001 |
Gleeson data set | ||||||
H1/H0 | 2 bb | 2 bb/bin | 2 bin | 1 bb | 1 bin | Post.Prob. |
2 bb | 0 | -2.173 | -3.390 | -2.065 | 6.705 | 0.066 |
2 bb/bin | 2.173 | 0 | -1.724 | 0.108 | 8.878 | 0.194 |
2 bin | 3.390 | 1.724 | 0 | 1.832 | 10.601 | 0.460 |
1 bb | 2.065 | -0.108 | -1.832 | 0 | 8.770 | 0.276 |
1 bin | -6.705 | -8.878 | -10.601 | -8.770 | 0 | 0.003 |
Hammoud data set | ||||||
H1/H0 | 2 bb | 2 bb/bin | 2 bin | 1 bb | 1 bin | Post.Prob. |
2 bb | 0 | -3.514 | -3.513 | -1.114 | 5.951 | 0.070 |
2 bb/bin | 3.514 | 0 | 0.020 | 2.400 | 9.465 | 0.404 |
2 bin | 3.513 | -0.020 | 0 | 2.380 | 9.444 | 0.400 |
1 bb | 1.114 | -2.400 | -2.380 | 0 | 7.064 | 0.122 |
1 bin | -5.951 | -9.465 | -7.064 | -7.064 | 0 | 0.004 |