| ModelStMoE-class {meteorits} | R Documentation |
A Reference Class which represents a fitted StMoE model.
Description
ModelStMoE represents an estimated StMoE model.
Fields
paramA ParamStMoE object. It contains the estimated values of the parameters.
statA StatStMoE object. It contains all the statistics associated to the StMoE model.
Methods
plot(what = c("meancurve", "confregions", "clusters", "loglikelihood"), ...)Plot method.
whatThe type of graph requested:
-
"meancurve" =Estimated mean and estimated experts means given the inputX(fieldsEyandEy_kof class StatStMoE). -
"confregions" =Estimated mean and confidence regions. Confidence regions are computed as plus and minus twice the estimated standard deviation (the squarre root of the fieldVaryof class StatStMoE). -
"clusters" =Estimated experts means (fieldEy_k) and hard partition (fieldklasof class StatStMoE). -
"loglikelihood" =Value of the log-likelihood for each iteration (fieldstored_loglikof class StatStMoE).
-
...Other graphics parameters.
By default, all the graphs mentioned above are produced.
summary(digits = getOption("digits"))Summary method.
digitsThe number of significant digits to use when printing.
See Also
Examples
data(tempanomalies)
x <- tempanomalies$Year
y <- tempanomalies$AnnualAnomaly
stmoe <- emStMoE(X = x, Y = y, K = 2, p = 1, threshold = 1e-4, verbose = TRUE)
# stmoe is a ModelSTMoE object. It contains some methods such as 'summary' and 'plot'
stmoe$summary()
stmoe$plot()
# stmoe has also two fields, stat and param which are reference classes as well
# Log-likelihood:
stmoe$stat$loglik
# Parameters of the polynomial regressions:
stmoe$param$beta