ModelTMoE-class: A Reference Class which represents a fitted TMoE model.

In fchamroukhi/tMoE: t Mixture-of-Experts: Robust Clustering of Complex Heterogeneous Non-Normal Data

Description

ModelTMoE represents an estimated TMoE model.

Fields

param

A ParamTMoE object. It contains the estimated values of the parameters.

stat

A StatTMoE object. It contains all the statistics associated to the TMoE model.

Methods

plot(what = c("meancurve", "confregions", "clusters", "loglikelihood"), ...)

Plot method.

what

The type of graph requested:

• "meancurve" = Estimated mean and estimated experts means given the input X (fields Ey and Ey_k of class StatTMoE).

• "confregions" = Estimated mean and confidence regions. Confidence regions are computed as plus and minus twice the estimated standard deviation (the squarre root of the field Vary of class StatTMoE).

• "clusters" = Estimated experts means (field Ey_k) and hard partition (field klas of class StatTMoE).

• "loglikelihood" = Value of the log-likelihood for each iteration (field stored_loglik of class StatTMoE).

...

Other graphics parameters.

By default, all the graphs mentioned above are produced.

summary(digits = getOption("digits"))

Summary method.

digits

The number of significant digits to use when printing.

See Also

ParamTMoE, StatTMoE

Examples

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

tmoe <- emTMoE(X = x, Y = y, K = 2, p = 1, verbose = TRUE)

# tmoe is a ModelTMoE object. It contains some methods such as 'summary' and 'plot'
tmoe$summary()
tmoe$plot()

# tmoe has also two fields, stat and param which are reference classes as well

# Log-likelihood:
tmoe$stat$loglik

# Parameters of the polynomial regressions:
tmoe$param$beta

fchamroukhi/tMoE documentation built on Sept. 23, 2019, 6:49 a.m.