Genetic Traits

The production traits are based on three lactations, and five regression coefficients per trait, as in the Canadian Test Day Model, with regressions on Legendre polynomials of days in milk. A lactation length of 365 days is assumed to be maximum. These allow the prediction of yields for each day in milk, and also allow for different shapes of lactation curves per animal. The predicted yields are expected amounts during an average day for a healthy, non-pregnant, non-stressed, animal. Production represents 105 different traits.

Each conformation trait is assumed to have a different genetic value in each of the first three lactations, but the three genetic values are correlated to each other at 0.80. This allows the genetic value for a trait to either improve or to decline with age. Thus, there are a total of 78 conformation traits.

There are eight fertility traits, and these are assumed to be genetically the same throughout the life of the animal, but with age and parity effects added for traits that observed more than once per cow. The traits are the same as those in the Canadian fertility trait evaluation system.

Weight and height are modelled by the Von Bertalanfy function which has three parameters, A, B, and K. This function describes weight and height at different ages throughout life. However, weights are affected by lactation, pregnancy, heat, cold, and diseases. These are additional effects that must be added or subtracted onto the expected weights. The effects differ depending on the parity number. Height is not very variable during the life of the animal, and once it reaches the mature value (i.e. the A parameter), then further change should not occur. Dry matter intake (DMI) is very similar across animals from birth to first calving, but there can still be individual variation. Body condition score, (BCS), is observed only within lactations. BCS changes during a lactation are assumed to be the same for all cows, just variation in the initial BCS score. Eventually BCS could be modelled with the production traits when parameters have been estimated.

These are other traits with single values. The total number of traits is 220.