Abstract
To Identify the best model to describe the lactation curve of dairy cows from the Western region of Santa Catarina State, brazil, among the various mathematical functions cited in literature. Milking control data (1.284) from 129 Holstein cows were used. Model parameters were estimated by non-linear regression through NLIN procedure (SAS System), using the Gauss-Newton method. The criteria for choosing the best model were: quasi-minimal re-sidual, mean absolute eviation, coefficient of determination (R²), best fit index, visual assessment of estimated and observed curves, and the residuals distribution graph. The model of Wood (1967) was the best to describe the lac-tation curve of cows, presenting R² of 0.95, DMA of 0.56, QMR of 27.57, the lowest index (4.86), the lowest residual range (between 0.5 and -0.5 liters) and an estimated curve more consistent with the observed.
Translated title of the contribution | Mathematical models to adjust the lactation curve of holstein cows raised in the Western region of Santa Catarina State, Brazil |
---|---|
Original language | Portuguese |
Pages (from-to) | 137-143 |
Journal | Revista Academica Ciencia Animal |
Volume | 11 |
Issue number | 2 |
DOIs | |
State | Published - Apr 1 2013 |
Externally published | Yes |
Keywords
- Non-linear models
- Projection of lactation
- Milk production
- Peak milk production