Mark-recapture data arises from studies in which animals are individually marked and released into a population, so that in subsequent time periods attempts may be made to recapture the animal.  The Cormack-Jolly-Seber model (Cormack, 1964; Jolly 1965; Seber 1965), when fitted to such data, allows survival probabilities, as well as recaptures probabilities to be estimated.  This simplest model has been extended in a number of ways; in particular, one influential extension was to model multi-site data.  The Arnason-Schwarz model (Arnason 1972, 1973; Schwarz et al., 1993) incorporates movement between sites using first-order Markovian transitions.   The Arnason-Schwarz model has been further promoted by Lebreton et al, (1999) to be a general framework for combined mark-recapture and recovery data.

Multi-state mark recapture model selection can be complex when using existing tools such as Akaike’s information criterion.  This software implements a step-wise model selection procedure using score tests (McCrea and Morgan, 2009).    The score test procedure is not hampered by the need to fit several inappropriate multi-parameter models that might result in likelihood surfaces with multiple optima, thus avoiding issues with other model selection approaches. 

Software MUSE (Model selection Using ScorE tests) performs step-wise score tests for product multinomial likelihoods and computes the expected information matrix for the construction of the score test statistic.  Options within the software allow the user to compare the performance of the observed and expected information matrix.  The software will also identify parameter redundant models.

References

• Arnason, A.N. (1972) Parameter estimates from mark-recapture-recovery experiments on two populations subject to migration and death. Res. Popul. Ecology 13, 91-113.
  
• Arnason, A.N. (1973) The estimation of populations size, migration rates, and survival in a stratified population. Res. Popul. Ecology 15, 1-8.
  
• Cormack, R.M. (1964) Estimates of survival from sightings of marked animals. Biometrika 51, 429-438.
  
• Jolly, G.M. (1965) Explicit estimates from capture-recapture data with both death and immigration stochastic models. Biometrika 90, 435-444.
  
• McCrea, R.S. and Morgan, B.J.T. (2009) Mark-recapture model selection using score tests. Submitted.
  
• Schwarz, C.G., Schweigert, J.F. and Arnason, A.N. (1993) Estimating migration rates using tag recovery data. Biometrics 59, 291-318.
  
• Seber, G.A.F. (1965) A note on the multiple-capture census. Biometrika 52, 249-259.