|
Statistical modelling
We can repeat the analysis for YLD210, saving the residuals in RES210 and the expected values (adding 323 kg to each residual) in EXP210.
Finally, by using Data → Calculations… and the 'ROUND' function, the respective total, residual and expected columns can be converted to whole numbers.
|
|
***** Analysis of variance *****
Variate: YLD210
Source of variation
|
d.f.
|
s.s.
|
m.s.
|
v.r.
|
HDGRP
|
7
|
1479516.
|
211359.
|
38.12
|
HDGRP.HERD
|
29
|
963397.
|
33221.
|
5.99
|
Residual
|
350
|
1940493.
|
5544.
|
|
Total
|
386
|
4383407.
|
|
|
* MESSAGE: the following units have large residuals.*
*units*
|
38
|
233.
|
s.e.
|
71.
|
*units*
|
81
|
-216.
|
s.e.
|
71.
|
*units*
|
159
|
241.
|
s.e.
|
71.
|
***** Tables of means *****
Variate: YLD210
Grand mean 323.
HDGRP
|
KANYAN
|
KASHON
|
KIKAAS
|
MASHA
|
|
|
412.
|
344.
|
256.
|
317.
|
rep.
|
83
|
62
|
43
|
11
|
HDGRP
|
MUTONT
|
RUHENG
|
RUSHER
|
RYERU
|
|
|
259.
|
227.
|
277.
|
342.
|
rep.
|
48
|
35
|
22
|
83
|
*** Standard errors of differences of means ***
Table
HDGRP
rep.
|
unequal
|
|
d.f.
|
350
|
|
s.e.d.
|
31.7X
|
min.rep
|
|
|
23.9
|
max-min
|
|
|
11.6X
|
max.rep
|
|
|
