Conceptual framework for investigating causal effects from observational data in livestock

Area of Science:

• Animal Science
• Agricultural Systems
• Causal Inference

Background:

• Efficient management of complex biological systems like animal agriculture relies on understanding causal mechanisms.
• Observational data from livestock operations offer opportunities for causal insight, despite inherent limitations.
• Recent advancements in causal inference provide theoretical and methodological foundations.

Purpose of the Study:

• Introduce a unifying conceptual framework for investigating causal effects from observational data in livestock.
• Illustrate the framework's implementation within animal science contexts.
• Discuss the opportunities and challenges of applying this framework.

Main Methods:

• Utilizing directed acyclic graphs (DAGs) as a foundational conceptual tool.
• Employing DAGs to encode structural mechanisms and probabilistic implications of causal models.
• Focusing on DAG elicitation and causal identification for observational data analysis.

Main Results:

• The review presents a structured approach to causal inference in animal agriculture.
• It highlights the utility of DAGs for modeling complex biological systems.
• Identifies key assumptions and limitations inherent in causal inference from observational data.

Conclusions:

• The proposed framework facilitates causal inference from observational data in animal sciences.
• Practical recommendations are provided for implementing causal inference methods.
• This approach enhances the understanding and management of livestock production systems.