Selecting the desirable method for predicting 305-day lactation yields in Mongolia

Area of Science:

• Animal Science
• Dairy Production
• Quantitative Genetics

Background:

• Mongolian dairy sector requires enhancement in milk quantity and quality.
• Accurate prediction of 305-day milk yield is crucial for genetic improvement and herd management.
• Existing prediction methods need evaluation for Mongolian dairy cows.

Purpose of the Study:

• To compare the accuracy of three methods for predicting 305-day milk yields in Mongolian cows.
• To identify the most suitable prediction model for optimizing dairy production in Mongolia.

Main Methods:

• Comparison of Test Interval Method (TIM), Multiple-Trait Prediction (MTP), and Random Regression Model (RRM).
• Utilized daily milk records (1986-2007) from Japan and monthly test-day records (1985-2005) from Hokkaido.
• Employed Wilmink's model for MTP and cubic Legendre polynomials with Wilmink's function parameters for RRM.

Main Results:

• The Random Regression Model (RRM) demonstrated superior accuracy in predicting 305-day milk yields.
• RRM's accuracy increased significantly when utilizing more than four data records.
• Estimates from RRM were found to be the most precise among the compared methods.

Conclusions:

• The Random Regression Model (RRM) is the most desirable method for predicting 305-day milk yields in Mongolia.
• Further research is needed to validate RRM's performance with Mongolian test-day records.
• Accurate yield prediction supports efficient breeding programs and dairy industry development.