emStMoE: emStMoE implements the ECM algorithm to fit a Skew-t Mixture...
In meteorits: Mixture-of-Experts Modeling for Complex Non-Normal Distributions

Description Usage Arguments Details Value See Also Examples

emStMoE implements the maximum-likelihood parameter estimation of a Skew-t Mixture of Experts (StMoE) model by the Expectation Conditional Maximization (ECM) algorithm.

1 2	emStMoE(X, Y, K, p = 3, q = 1, n_tries = 1, max_iter = 1500, threshold = 1e-06, verbose = FALSE, verbose_IRLS = FALSE)

`X`	Numeric vector of length n representing the covariates/inputs x_{1},…,x_{n}.
`Y`	Numeric vector of length n representing the observed response/output y_{1},…,y_{n}.
`K`	The number of experts.
`p`	Optional. The order of the polynomial regression for the experts.
`q`	Optional. The order of the logistic regression for the gating network.
`n_tries`	Optional. Number of runs of the ECM algorithm. The solution providing the highest log-likelihood will be returned.
`max_iter`	Optional. The maximum number of iterations for the ECM algorithm.
`threshold`	Optional. A numeric value specifying the threshold for the relative difference of log-likelihood between two steps of the ECM as stopping criteria.
`verbose`	Optional. A logical value indicating whether or not values of the log-likelihood should be printed during ECM iterations.
`verbose_IRLS`	Optional. A logical value indicating whether or not values of the criterion optimized by IRLS should be printed at each step of the ECM algorithm.

emStMoE function implements the ECM algorithm for the StMoE model. This function starts with an initialization of the parameters done by the method initParam of the class ParamStMoE, then it alternates between the E-Step (method of the class StatStMoE) and the M-Step (method of the class ParamStMoE) until convergence (until the relative variation of log-likelihood between two steps of the ECM algorithm is less than the threshold parameter).

ECM returns an object of class ModelStMoE.

ModelStMoE, ParamStMoE, StatStMoE

data(tempanomalies)
x <- tempanomalies$Year
y <- tempanomalies$AnnualAnomaly

stmoe <- emStMoE(X = x, Y = y, K = 2, p = 1, threshold = 1e-4, verbose = TRUE)

stmoe$summary()

stmoe$plot()