作者: Shinichi Nakagawa , Robert P. Freckleton
DOI: 10.1016/J.TREE.2008.06.014
关键词:
摘要: The most common approach to dealing with missing data is delete cases containing observations. However, this reduces statistical power and increases estimation bias. A recent study shows how estimates of heritability selection can be biased when the 'invisible fraction' (missing due mortality) ignored, thus demonstrating dangers neglecting in ecology evolution. We highlight advances procedures handling their relevance applicability.