Agriculture Reference
In-Depth Information
technique (Fishman, 1978). As an example,
the NRC (2001) model of dairy cattle calcu-
lates the net energy for lactation (
NE
L
) of
feedstuffs based on their chemical compos-
ition in dry matter, crude protein, neutral
detergent fibre, lignin, ether extracts, ash,
neutral detergent insoluble protein and acid
detergent insoluble protein. The chemical
composition of a given feedstuff does vary.
Knowing the variances and covariances
among the chemical elements of a given
feedstuff allows the estimation of the distri-
the frequency distribution of
NE
L
concentra-
tion for two feeds commonly used in dairy
cattle feeding: soybean meal and distiller's
dried grains with solubles (DDGS). It is obvi-
ous that the biggest difference between the
energy concentration of soybean meal and
DDGS is not in their mean energy contents,
which are nearly identical, but in the vari-
ability of their actual energy content. Based
on energy, a deterministic approach would
conclude that animal performance should
be equal regardless of which one of the two
sources is used. From a stochastic viewpoint,
one would expect animal performance to be
much more variable with DDGS than with
soybean meal-based diets.
The last source of error entering Eqn 5.4
is the error in the functional form. Regard-
less of whether the model is a simple linear
regression or a complex set of differential
equations, models are always simplifications
of the true but very complex world that is
being investigated. For any given phenom-
ena or system being modelled, there exists a
nearly infinite set of possible functional forms
14
13
12
Soybean meal
11
10
9
8
7
6
5
4
3
2
Distiller's dried
grains with solubles
1
0
1.6
1.65
1.7
1.75
1.8
1.85
1.9
1.95
2
2.05
2.1
2.15
2.2
2.25
2.3
2.35
2.4
Net energy for lactation (Mcal/kg)
Fig. 5.3.
Frequency distribution of net energy for lactation (
NE
L
) concentration of soybean meal and distiller's
dried grains with solubles based on a Monte Carlo simulation. (From Table
15-1
of NRC, 2001.)
