Some suggest that BW and top-view body area have a linear
relationship (Marchant et al., 1999; Schofield et al., 1999; White
et al., 2004) and use a single linear regression equation to estimate
the live BW of animals from the body area based on the interpretation
of individual images. Schofield et al. (1999) suggested that
different breeds may require different algorithms for BW prediction.
Also Fisher et al. (2003) and Green et al. (2003) suggested a
need for unique algorithms for specific breeds or lines of pigs. More
recently, researchers have been highlighting the benefits of mixed
effects models (Schinkel et al., 2009) and justify their argument
that mixed effects model is easily adaptable to stochastic modelling.
However, despite the advantages of mixed effects models
compared to fixed effects models, it is important to note that there
is a large amount of variation in the accuracy of different mixed
effects models.