Skip to main content

Table 5 Actual, refitted, and recovered variances of regression coefficients in Eqs. 1–7 in Table 8

From: Model errors in tree biomass estimates computed with an approximation to a missing covariance matrix

Eq
#
Actual Refitted Recovered Recovered
(robust)
P 1
×100
P 2
×100
P 3
×100
1 (0.04, 0.01) (0.08, 0.01) (0.07, 0.01) (0.10, 0.02) 56 74 49
2 (0.55, 0.42, 0.020) (2.64, 1.39, 0.044) (2.36, 1.25, 0.039) (2.88, 1.52, 0.048) 0.2 0.1 0.1
3 (0.27, 0.048) (0.29, 0.041) (0.26, 0.039) (0.35, 0.053) 99 98 96
4 (0.084, 0.017) (0.33, 0.049) (0.29, 0.043) (0.37, 0.06) 2 5 1
5 n.a. (0.32, 7.1)10−3 (0.20, 9.4)10−3 (0.23, 10.6)10−3 n.a. n.a. n.a.
6 n.a. (0.61, 4.7)10−1 (0.42, 32.2)10−2 (0.12, 9.0)10−1 n.a. n.a. n.a.
7 n.a. (0.98, 16.8)10−3 (0.29, 19.4)10−3 (0.74, 29.4)10−3 n.a. n.a. n.a.
  1. Actual covariance matrices are based on sample sizes listed in Table 8. P i (i = 1, 2, 3) is the probability under H 0 of: (1) actual = refitted; (2) actual = recovered; (3) actual = robust.