Browsing by Subject "Statistical models"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Restricted
    S24-1 Progress with the statistical analysis of primary molt
    (Current Zoology, 2006) Underhill, Les G; Serra, Lorenzo; Brandão, Anabela
    This paper considers two extensions to the statistical analysis of primary molt. The first extension relates to the development of statistical models that allow the simultaneous estimation of molt parameters for diverse groups of birds. Parameter estimation uses the method of maximum likelihood and the EM algorithm. The likelihood ratio criterion is applied in statistical hypothesis-testing for equality of sets of parameters, and the Akaike Information Criterion is used to guide model selection. This extension is illustrated using primary molt data for willow warblers (Pylloscopus trochilus) over 11 years. The second extension is the realization that important insights into the way that primary molt is incorporated into the annual cycle of birds can be obtained by investigating the parameters of molt of individual primary feathers. This extension is illustrated using six data sets from four species of waders, Charadriiformes.