Split two randomly selected clusters based on a model with one component smaller than the current one. This procedure is repeated within a small-EM scheme. The best split is chose to initialize the model.
splitEM_GLM(y, x, K, smallerModel, tsplit = 10, maxIter = 20,
emthreshold = 1e-08, maxNR = 5, nCores,
split = TRUE, R0, method)val
y
x
k
smla
tsp
max
thr
maxn
nc
spi
ro
meth
Panagiotis Papastamoulis
Papastamoulis, P., Martin-Magniette, M. L., and Maugis-Rabusseau, C. (2016). On the estimation of mixtures of Poisson regression models with large number of components. Computational Statistics & Data Analysis, 93, 97-106.