Statistical modelling
We now change the order in which the effects are fitted so
that GENOTYPE is added last. For the purposes of this output GenStat has also
provided the predicted least squares means for each genotype (obtained by
clicking the Predict button that appears after the model has been fitted
)
together with their standard errors.
Note that the parameter estimates remain the same
regardless of the order in which the terms are added to the model. However,
compairing
this output with the one given earlier in which the term GENOTYPE was fitted
first, the mean square accounted for by GENOTYPE, after correcting for all the
other terms in the model, is reduced from 190.142 to 25.286 kg2. This
demonstrates the impact of the partial confounding of other factors,
particularly year, on the variation that can be attributed uniquely to genotype.
Genotype differences, nevertheless, remain significant (P < 0.01)
|
|
Response variate: WEANWT
Fitted terms: Constant + YEAR + SEX + AGEWEAN + DL
+ DQ + GENOTYPE
*** Estimates of parameters ***
|
estimate |
s.e. |
t(687) |
Constant |
0.25 |
1.07 |
0.23 |
YEAR 92 |
-1.565 |
0.293 |
-5.34 |
YEAR 93 |
-1.099 |
0.276 |
-3.99 |
YEAR 94 |
-2.820 |
0.359 |
-7.85 |
YEAR 95 |
-3.215 |
0.346 |
-9.30 |
YEAR 96 |
-2.342 |
0.391 |
-5.99 |
SEX M |
0.476 |
0.170 |
2.81 |
AGEWEAN |
0.07026 |
0.00886 |
7.93 |
DL |
2.725 |
0.315 |
8.65 |
DQ |
-0.269 |
0.034 |
-7.90 |
GENOTYPE RD |
-0.389 |
0.220 |
-1.77 |
GENOTYPE RR |
-1.040 |
0.271 |
-3.84 |
GENOTYPE DR |
-0.511 |
0.304 |
-1.68 |
*** Accumulated analysis of variance ***
Change |
d.f. |
s.s. |
m.s. |
v.r. |
+YEAR |
5 |
1208.149 |
241.630 |
48.92 |
+SEX |
1 |
55.983 |
55.983 |
11.34 |
+AGEWEAN |
1 |
344.206 |
344.206 |
69.69 |
+DL |
1 |
151.513 |
151.513 |
30.68 |
+DQ |
1 |
275.795 |
275.795 |
55.84 |
+GENOTYPE |
3 |
75.857 |
25.286 |
5.12 |
Residual |
687 |
3392.947 |
4.939 |
|
Total |
699 |
5504.450 |
7.875 |
|
GENOTYPE |
Prediction |
s.e. |
DD |
11.552 |
0.159 |
DR |
11.041 |
0.240 |
RD |
11.163 |
0.176 |
RR |
10.512 |
0.193 |
|
|
|
|