Hartling, Joey2019-08-302019-08-302019-08-30http://hdl.handle.net/10222/76368State space models require the ability to perform filtering, smoothing and prediction during analysis. To perform these procedures fairly complex computational algorithms are required. There is a consequent need for software tools to facilitate the implementation of state space models. One of leading choices for computation data analysis is the statistical programming language, R. And a key area where analysis tools are lacking is for advanced state space models. This thesis outlines the development of a comprehensive toolkit for nonlinear and non-Gaussian state space models: the nLnG package.enstate space modelstime series analysisfilteringsmoothingparameter estimationKalman filterKalman smootherparticle filterparticle smootherstate augmentationmultiple iterative filteringrandom walk modelpackage developmentRprogramminganimal tracking datanonlinearnon-Gaussianalgorithms