Abstract:
In this paper, we suggest an approach to estimate the probability density function for bivariate circular data. Kernel density estimation technique with different concentration parameters and different kernel functions are used to construct this approach. In our approach, we used Wrapped Cauchy distribution and Wrapped Normal distribution as a kernel function. Moreover, the procedure that we have suggested is applied to two types of data (simulated data and protein data). Firstly, we apply the approach that we have suggested to simulated data, which are generated from the bivariate von-Mises distribution, to highlight the joint probability density function for the bivariate von-Mises distribution. Secondly, the proposed approach is applied to real data, from the Protein Data Bank (PDB), to show the joint probability density function for some types of proteins.
|