Table 1.

Random regression sire variance estimates of the Legendre polynomial intercept (I, k = 1) and slope (S, k = 2), covariance (IxS) and residual variance estimates for different classes (p from 1 to 5) in different models (UM, RRMw, RRMg, RRMw-m, RRMw-g, RRMITw-m, RRMITg-m). The approximate standard errors are shown below each parameter.

	Intercept (I) (k = 1)	Slope (S) (k = 2)	I x S	$${\widehat{\sigma }}_{e|p=1}^2$$	$${\widehat{\sigma }}_{e|p=2}^2$$	$${\widehat{\sigma }}_{e|p=3}^2$$	$${\widehat{\sigma }}_{e|p=4}^2$$	$${\widehat{\sigma }}_{e|p=5}^2$$
UM	80.6			629.2
	5.6			6.9
RRMw	66.9	19.3	14.4	478.1	562.4	590.3	664.0	738.84
	6.2	4.7	3.8	11.8	10.1	15.5	26.9	40.6
RRMg	72.0	18.3	16.6	530.1	575.5	636.6	628.4	762.8
	6.8	5.3	4.1	12.6	9.8	13.8	22.3	43.8
RRMw-m	71.9	14.5	12.8	479.4	553.8	614.9	657.1	763.7
	6.1	3.9	3.5	10.3	9.5	14.4	24.0	43.1
RRMg-m	81.9	11.2	9.6	523.2	592.8	621.9	630.3	750.5
	6.6	3.3	3.0	9.5	9.2	11.8	16.4	33.4
RRMITw-m	77.2	16.6	16.6	476.1	564.6	617.3	681.5	854.0
	6.9	4.4	4.6	9.4	9.2	16.1	29.4	55.6
RRMITg-m	81.3	12.5	20.8	483.6	575.6	604.2	671.8	839.0
	6.5	4.1	4.7	9.4	8.8	13.4	21.4	42.7