Goestica-class function

Class "Goestica": GO-GARCH models estimated by fast ICA

This class contains the GoGARCH class and has the mixing matrix $A$ as additional slot. latin1

class

Objects from the Class

Objects can be created by calls of the form new("Goestmm", ...), or with the function gogarch whereby method = "ica" has been set.

Slots

• ica:: Object of class "list": List object returned by fastICA.
• Z:: Object of class "matrix": Transformation matrix.
• U:: Object of class "matrix": Orthogonal matrix.
• Y:: Object of class "matrix": Extracted component matrix.
• H:: Object of class "list": List of conditional variance/covariance matrices.
• models:: Object of class "list": List of univariate GARCH model fits.
• estby:: Object of class "character": Estimation method.
• X:: Object of class "matrix": The data matrix.
• V:: Object of class "matrix": Covariance matrix of X.
• P:: Object of class "matrix": Left singular values of Var/Cov matrix of X.
• Dsqr:: Object of class "matrix": Square roots of eigenvalues on diagonal, else zero.
• garchf:: Object of class "formula": Garch formula used for uncorrelated component GARCH models.
• name:: Object of class "character": The name of the original data object.

Extends

Class "GoGARCH", directly. Class "Goinit", by class "GoGARCH", distance 2.

Methods

• cvar: Returns the conditional variances as object with class attribute "mts" "ts".
• ccov: Returns the conditional co-variances as object with class attribute "mts" "ts".
• ccor: Returns the conditional correlationsas object with class attribute "mts" "ts".
• coef: Returns the coeffiecients of the component GARCH models.
• converged: Returns the convergence codes of the component GARCH models.
• formula: Returns the formula for the component GARCH models.
• goest: Fast ICA estimation of Go-GARCH models.
• plot: Plotting of the conditional correlations.
• predict: Returns the conditional covariances and mean forecasts and the forecasts of the component GARCH models, object is of class Gopredict.
• residuals: Returns the residuals of the Go-GARCH model as object with class attribute "mts" "ts".
• resid: Returns the residuals of the Go-GARCH model as object with class attribute "mts" "ts".
• show: show-method for objects of class Goestmm.
• summary: summary-method for objects of class Goestml, object is of class Gosum.
• update: Updates an object of class Goestml.

References

Broda, S.A. and Paolella, M.S. (2008): CHICAGO: A Fast and Accurate Method for Portfolio Risk Calculation, Swiss Finance Institute, Research Paper Series No. 08-08, Zuerich.

Author(s)

Bernhard Pfaff

See Also

GoGARCH, Goinit, Gosum, Gopredict, goest-methods and gogarch