summary.ClusterVAR {ClusterVAR}

Summary of ClusterVAR objects

Description

Takes the output of the LCVAR function and creates a small summary of the fitted model(s).

Usage

## S3 method for class 'ClusterVAR'
summary(object,
  show = "BPC",
  TS_criterion = "SC",
  global_criterion = "BIC",
  Number_of_Clusters = NULL,
  Number_of_Lags = NULL,
  ...)

Arguments

object

An output object of the LCVAR function.

show

Indicate how models should be summarized, the possible choices are "BPC", "GNC"and "GNL". show = "BPC" compares models with different time lags for each number of clusters. If show = "BPC", for each number of clusters the model with the best lag is selected and displayed in the output. The best lag is selected through the time-series information criterion specified with the argument TS_criterion (see below). show = "GNC" shows all models with different lags for a given number of clusters, this number of clusters is specified through Number_of_Clusters (see below). show = "GNL" shows for each number of clusters the model where all lags are fixed to a given number, this number of lags is specified through Number_of_Lags (see below). Out of these models, the best model in terms of the number of clusters is selected by the information criterion selected with the argument global_criterion (see below). Defaults to show = "BPC".

TS_criterion

The information criterion to select the best model between models with a different number of lags but with the same number of clusters. The possible choices are "SC" and "HQ". Defaults to TS_criterion = "SC".

global_criterion

The information criterion to select the best model between models with different numbers of clusters but with the same number of lags. The possible choices are "BIC" and "ICL". Defaults to global_criterion = "BIC".

Number_of_Clusters

An integer. Specify the fixed number of clusters when using show = "GNC". Defaults to Number_of_Clusters = NULL.

Number_of_Lags

An integer. Specify the fixed number of lags when using show = "GNL". Defaults to Number_of_Lags = the lowest number of lags specified in object.

...

Pass additional arguments.

Value

FunctionOutput

Is a data frame containing summaries of the fitted models.

Author(s)

Anja Ernst & Jonas Haslbeck

References

Hamilton, J. (1994), Time Series Analysis, Princeton University Press, Princeton.

Hannan, E. J. and B. G. Quinn (1979), The determination of the order of an autoregression, Journal of the Royal Statistical Society.

Lütkepohl, H. (2006), New Introduction to Multiple Time Series Analysis, Springer, New York.

Quinn, B. (1980), Order determination for a multivariate autoregression, Journal of the Royal Statistical Society.

Biernacki, C., Celeux, G., & Govaert, G. (2000). Assessing a mixture model for clustering with the integrated completed likelihood. IEEE Transactions on Pattern Analysis and Machine Intelligence.

Schwarz, G. (1978). Estimating the Dimension of a Model. The Annals of Statistics.

See Also

plot.ClusterVAR(), coef.ClusterVAR()

Examples

LCVAR_outExample <- LCVAR(Data = ExampleData,
                          yVars = 1:4, ID = 5, Beep = 6,
                          xContinuous = 7, xFactor = 8,
                          Clusters = 1:2, Lags = 1:2,
                          Cores = 2, RndSeed = 3,
                          Rand = 2, it = 25)

summary(LCVAR_outExample)
summary(object = LCVAR_outExample, show = "GNL", Number_of_Lags = 1)
summary(object = LCVAR_outExample, show = "GNL", Number_of_Lags = 1, global_criterion = "ICL")
summary(object = LCVAR_outExample, show = "GNC", Number_of_Clusters = 2)
summary(object = LCVAR_outExample, show = "GNC", Number_of_Clusters = 2, TS_criterion = "HQ")