BIC to choose the number of components in a model based clustering using mixtures of rotationally symmetric distributions
bic.mixvmf(x, G = 5, n.start = , tol = 1e-6, maxiters = 500)
bic.mixspcauchy(x, G = 5, n.start = 5, tol = 1e-6, maxiters = 500)
bic.mixpkbd(x, G = 5, n.start = 5, tol = 1e-6, maxiters = 500)A plot of the BIC values and a list including:
The BIC values for all the models tested.
The ICL values for all the models tested.
The run time of the algorithm. A numeric vector. The first element is the user time, the second element is the system time and the third element is the elapsed time.
A matrix containing directional data.
The maximum number of clusters to be tested. Default value is 5.
The number of random starts to try. See also R's built-in function kmeans for more information about this.
The tolerance value to terminate the EM algorithm.
The maximum number of iterations to perform.
Michail Tsagris.
R implementation and documentation: Michail Tsagris mtsagris@uoc.gr.
The function computes the BIC (and ICL) to decide on the optimal number of clusters when using mixtures of von Mises-Fisher, mixtures of spherical Cauchy or mixtures of Poisson kernel-based distributions.
Hornik, K. and Grun, B. (2014). movMF: An R package for fitting mixtures of von Mises-Fisher distributions. Journal of Statistical Software, 58(10): 1--31.
Biernacki C., Celeux G. and Govaert, G. (2000). Assessing a mixture model for clustering with the integrated completed likelihood. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(7): 719--725.
Tsagris M., Papastamoulis P. and Kato S. (2024). Directional data analysis using the spherical Cauchy and the Poisson kernel-based distribution. https://arxiv.org/pdf/2409.03292
mixvmf.mle, rmixvmf, mixvmf.contour
x <- as.matrix( iris[, 1:4] )
x <- x / sqrt( rowSums(x^2) )
bic.mixvmf(x)
Run the code above in your browser using DataLab