TABLE 2.
Performance of our conditional-likelihood approach at estimating σ for the spatial model
| β = 1
|
β = 2
|
||||||
|---|---|---|---|---|---|---|---|
| m | θ |  |
RMSE | Coverage (%) |  |
RMSE | Coverage (%) |
| 5 | 2 | 0.99 | 0.45 | 95 | 1.00 | 0.42 | 96 |
| 5 | 5 | 1.09 | 0.46 | 93 | 0.99 | 0.38 | 94 |
| 10 | 5 | 1.02 | 0.28 | 95 | 1.04 | 0.29 | 95 |
| 10 | 15 | 1.05 | 0.24 | 94 | 1.03 | 0.23 | 94 |
| 20 | 15 | 1.13 | 0.22 | 83 | 1.18 | 0.26 | 79 |
| 20 | 30 | 1.14 | 0.27 | 79 | 1.20 | 0.29 | 73 |
| 40 | 15 | 1.22 | 0.31 | 57 | 1.23 | 0.30 | 51 |
| 40 | 30 | 1.22 | 0.30 | 45 | 1.28 | 0.32 | 40 |
We report the mean of the estimates of σ (truth σ = 1), the root mean square error of the estimates, and the coverage probability of 95% approximate confidence intervals. (The grid of σ-values ranged from 0 to 4 for m = 5 and m = 10 and from 0 to 2 for m > 10.)