Uninformative Parameters and Model Selection Using Akaike's Information Criterion

摘要: Abstract As use of Akaike's Information Criterion (AIC) for model selection has become increasingly common, so a mistake involving interpretation models that are within 2 AIC units (ΔAIC ≤ 2) the top-supported model. Such <2 ΔAIC because penalty one additional parameter is +2 units, but deviance not reduced by an amount sufficient to overcome 2-unit and, hence, provides no net reduction in AIC. Simply put, uninformative does explain enough variation justify its inclusion and it should be interpreted as having any ecological effect. Models with parameters frequently presented being competitive Journal Wildlife Management, including 72% all AIC-based papers 2008, authors readers need more aware this problem take appropriate steps eliminate misinterpretation. I reviewed 5 potential solutions problem: 1) report model...