ASAS-NANP symposium: mathematical modeling in animal nutrition: review of time series techniques and machine learning models for advancing livestock production

This is the authors' abstract. We don't add key points for this paper.

Correlated data, including time series where observations depend on previous values, frequently occur in various livestock production studies. There are many statistical models and machine learning algorithms that have been developed to analyze this type of data. This paper provides a review of these models. Specifically, a detailed overview of the AutoRegressive Integrated Moving Average with eXogenous inputs (ARIMAX) model, Dynamic Models and Kalman Filtering (KF), Deep Neural Network (DNN), Recurrent Neural Network (RNN), and Transformer models are provided. This review focuses on steps to train different statistical and machine learning models and lists the advantages and limitations of each model.