1: \begin{abstract}

2: This paper proposes a statistical analysis of movement data in ecology using partially observed stochastic differential equations. 

3: Usually, in movement ecology, parameters of these models are estimated using approximate 

4: maximum likelihood procedures based on the Euler-Maruyama discretization.

5: However, GPS sampling rate in ecology might not be large 

6: enough to ensure the stability and convergence of the Euler based estimates. To 

7: our best knowledge, there is no practical study to assess the performance of the 

8: Euler Maruyama method to estimate movement ecology models compared to other 

9: inference procedures for stochastic differential equations. In this paper, we 

10: propose such a practical study by comparing the Euler method with the Ozaki 

11: linearization method of \cite{ozaki:1992,shoji:1998}, an adaptive Kessler high order 

12: Gaussian approximation method introduced in \cite{kessler:1997,kessler:lindner:sorensen:2012,uchida:yoshida:2012}, and the Exact Algorithm 

13: based Monte Carlo Expectation Maximization approach of 

14: \cite{beskos:papaspiliopoulos:roberts:2006}. The performance of these methods 

15: are assessed using a new potential based  stochastic differential equation where 

16: the drift is given as the gradient of a mixture of attractive zones, which are 

17: of main interest in ecology and fisheries science. It is shown both on 

18: simulated data and actual fishing vessels data that the Euler method performs worse 

19: than the other procedures for non high frequency sampling schemes. We also show, 

20: on this model, that Kessler and Ozaki methods are quite robust and perform in a 

21: similar way as the exact method.

22: \end{abstract}

23: