EStepVar {cemco}

Calculate the E step of the CemCO algorithm with covariates effects on distributions means and distributions covariance matrices.

Description

Implements the expectation step of EM algorithm for parameterized Gaussian mixture models with covariates effects on the distribution means and the distribution covariance matrices. It is also used to calculate the posteriori probability of each observation belong to each cluster.

Usage

EStepVar(data, Y, phi, G, y_cov)

Arguments

data

A numeric vector, matrix, or data frame of observations. Non-numerical values should be converted to integer or float (e.g. dummies). If matrix or data frame, rows and columns correspond to observations (n) and variables (P).

Y

numeric matrix of data to use as covariates. Non-numerical values should be converted to integer or float (e.g. dummies).

phi

list of fitted parameters in the same format as the output of the CemCO function

G

An integer specifying the numbers of mixture components (clusters)

y_cov

numeric matrix of data to use as covariates for the covariance effect. Non-numerical values should be converted to integer or float (e.g. dummies).

Details

Calculate the a posteriori probability of each observation belong to each cluster given the data and the current parameters estimation.

Value

Returns a n x G numeric matrix where n represents the number of observations (number of rows of data) and G (the number of clusters). The value i, j represents the probability of the i-th observation belong to j-th cluster.

Author(s)

Relvas, C. & Fujita, A.

References

Stage I non-small cell lung cancer stratification by using a model-based clustering algorithm with covariates, Relvas et al.

Examples

set.seed(42)
X = cbind(rnorm(10), rnorm(10))
Y = cbind(rnorm(10), rnorm(10))
K = 2

fit <- CemCOVar(X, Y, K, Y[,1], max_iter=2, n_start=1, cores=1)
prob <- EStepVar(X, Y, fit[[1]], K, Y[,1])