Table 2 Sensitivity results for Medium scenario: estimation performance for \(N=50\) observations, \(\dot{z}_i =\) 10,000 reads, and \(T=50\) compositional elements at varying levels of misclassification
Hyperparameter | \(\beta _{\eta _{t1}} \ne 0\) | \(\beta _{\eta _{t1}} = 0\) | \(\pi _t(\varvec{x}_{ip})\) | |||
|---|---|---|---|---|---|---|
Setting | ABS | COV | ABS | COV | ABS | COV |
\(\nu _{tt} = 10\) | 0.466 | 0.893 | 0.302 | 0.947 | 0.630 | 0.560 |
\(\nu _{tt} = 1000\) | 0.904 | 0.793 | 0.446 | 0.982 | 0.623 | 0.559 |
\(\sigma ^2_{\eta } = \sigma ^2_{\gamma }= 1\) | 0.440 | 0.853 | 0.268 | 0.954 | 0.620 | 0.554 |
\(\sigma ^2_{\eta } = \sigma ^2_{\gamma }= 10\) | 0.514 | 0.877 | 0.330 | 0.930 | 0.629 | 0.569 |