### Tables 2 and 3 list the asymptotic e ciencies of the elements of the MCD estimator at some Student distributions F with degrees of freedom, for several values of p and for = 0:25 and 0:5: Corresponding values for the normal distribution (which is the limit case for ! 1) are also reported. One sees that at these heavier tailed distributions it remains true that the e ciency gain taking = 0:25 instead of the usual = 0:5 is considerable. Moreover, at the Student distribution with 5 degrees of freedom, the MCD (with = 0:25) is even more e cient than the classical covariance matrix estimator.

1999

Cited by 11

### Table 4: Estimates with Covariates

2004

### (Table 4). The result was a negative covariance between

"... In PAGE 6: ... Estimates of the covariance D1 were negative for environments were examined. Upper and lower bounds of 95% every trait (Table4 ). All five genotype H11003 environment confidencelimits werecomputed forpanmictic populationmeans interaction components were significant for grain yield.... In PAGE 6: ... were generally small with respect to main effects, and usually not larger than two standard errors except H92682AE RESULTS and D*2E for grain moisture and H*E for ear height (Ta- Inbreeding Depression ble 4). Predicted variances among lines in each genera- tion of inbreeding showed an increasing trend from the Inbreeding depression was found for all six traits in S0 (noninbred half-sib families) generation to the S4 both the evaluation of individual lines and in the bulk generation (Table4 ). The predicted variance among S4 entry experiment (Table 3).... In PAGE 6: ...etween 0.95 and 1.18 for remaining traits, demonstra- tween experiments except for grain yield (Table 3). The ting that inbreeding did not result in a doubling of total confidence interval on the rate of inbreeding depression genetic variance as expected under an additive model for grain yield in the evaluation of individual lines was (Table4 ). The ratio of total genetic variance at F H11005 1 less than half the size of the corresponding interval from to additive variance, also expected to be 2 under an the bulk entry experiment (Table 3).... In PAGE 6: ...39 and 2.28 for grain yield and grain moisture, respectively and was less than two for Genetic Variances other traits (Table4 ). In contrast to changes in total All five genotypic covariance components (H92682A, H92682 D, variance, large increases in variance of dominance devi- D1, D*2 , H*) were larger than two standard errors for ations were observed with inbreeding.... In PAGE 6: ....65, 3.33, and 3.01 times the variance of dominance deviations of noninbred individuals, H92682 D, for grain yield, ear height, and plant height, respectively. Estimates of the degree of dominance were over 2 for all traits except grain moisture (Table4 ). The degree of dominance for grain moisture was not significantly greater than 1.... In PAGE 6: ... The degree of dominance for grain moisture was not significantly greater than 1.0, which corresponds to complete domi- nance (Table4 ). Correlations between genotypic values, G, and breeding values, A, ranged from 0.... In PAGE 9: ... Conversely, a large H*, as we obtained, may suggest a few loci with large effects on Variances of breeding values and dominance devia- inbreeding depression, or high levels of linkage disequi- tions both increased with inbreeding: (i) variance of breed- librium so that alleles at sets of linked loci are acting ing values of inbred individuals is twice the variance of as single loci. Therefore, our large estimates of H* and breeding values of noninbred individuals by definition, the degree of dominance could suggest the presence of (ii) variance of inbred dominance deviations was greater a few regions with segregating recessives at several loci than the variance of panmictic dominance deviations tightly linked in repulsion phase with relatively large for five of six traits (Table4 ). However, the variance of effects.... ..."

### TABLE I PERFORMANCE OF THE COVARIATION ESTIMATOR.

2006

Cited by 5

### TABLE I PERFORMANCE OF THE COVARIATION ESTIMATOR.

### Table 5 With Income as an active covariate:

2003

Cited by 1

### Table 3. Standardized likelihood estimates for covariates

"... In PAGE 12: ...-.036 versus -.023). The stronger effects for the second interval in Beijing may be partly a consequence of a longer lag period, which provides a longer duration for effects to occur. Standardized estimates of the effects of covariates on functional limitation and economic well-being are shown in Table3 . The direction and significance for the majority of variables are as would be expected and, for the most part, effects across settings are similar.... ..."