Nothing Special   »   [go: up one dir, main page]
Scribd Logo
Upload

Welcome to Scribd!

  • 53 views

    Garchoxfit R W

    Uploaded by

    datsno
    This function defines a "garchOxFit" function that estimates EGARCH models using the Ox software. It takes in a formula for the mean and variance, the time series data, and other options. It writes the parameters, series, runs the Ox program, and returns an object with the residuals, conditional variances, coefficients, and other details of the EGARCH model fit.

    Copyright:

    © All Rights Reserved
    Download as TXT, PDF, TXT or read online from Scribd

    Garchoxfit R W

    Uploaded by

    datsno
    53 views2 pages
    This function defines a "garchOxFit" function that estimates EGARCH models using the Ox software. It takes in a formula for the mean and variance, the time series data, and other options. It writes the parameters, series, runs the Ox program, and returns an object with the residuals, conditional variances, coefficients, and other details of the EGARCH model fit.

    Original Description:

    garch

    Original Title

    garchoxfit_R_w

    Copyright

    © © All Rights Reserved

    Available Formats

    TXT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This function defines a "garchOxFit" function that estimates EGARCH models using the Ox software. It takes in a formula for the mean and variance, the time series data, and other options. It writes the parameters, series, runs the Ox program, and returns an object with the residuals, conditional variances, coefficients, and other details of the EGARCH model fit.

    Copyright:

    © All Rights Reserved
    Download as TXT, PDF, TXT or read online from Scribd
    Download as txt, pdf, or txt
    53 views2 pages

    Garchoxfit R W

    Uploaded by

    datsno
    This function defines a "garchOxFit" function that estimates EGARCH models using the Ox software. It takes in a formula for the mean and variance, the time series data, and other options. It writes the parameters, series, runs the Ox program, and returns an object with the residuals, conditional variances, coefficients, and other details of the EGARCH model fit.

    Copyright:

    © All Rights Reserved
    Download as TXT, PDF, TXT or read online from Scribd
    Download as txt, pdf, or txt
    of 2

    #### FTS 2007 (5/9/2007) created mainly to estimate EGARCH models in R.

    "garchOxFit" <-
    function (formula.mean = ~arma(0, 0), formula.var = ~garch(1, 1),
    series = x, cond.dist = c("gaussian", "t", "ged", "skewed-t"),
    include.mean = TRUE, include.var=TRUE, truncation = 100,
    trace = TRUE, title = NULL, arch.in.mean=0)
    {
    fit = list()
    fit$x = series
    # include.var = TRUE
    fit$csts = c(include.mean, include.var)
    distris = 0:3
    names(distris) = c("gaussian", "t", "ged", "skewed-t")
    distri = distris[cond.dist[1]]
    fit$cond.dist = cond.dist[1]
    if (missing(formula.mean)) {
    fit$formula.mean = ~arma(0, 0)
    fit$arma.orders = c(0, 0)
    }
    else {
    fit$arma.orders =
    as.numeric(strsplit(strsplit(strsplit(as.character(formula.mean),
    "\\(")[[2]][2], "\\)")[[1]], ",")[[1]])
    }
    arfima = FALSE
    fit$arfima = as.integer(arfima)
    if (missing(formula.var)) {
    fit$formula.var = ~garch(1, 1)
    fit$garch.orders = c(1, 1)
    }
    else {
    fit$garch.orders =
    as.numeric(strsplit(strsplit(strsplit(as.character(formula.var),
    "\\(")[[2]][2], "\\)")[[1]], ",")[[1]])
    }
    # arch.in.mean = 0
    fit$arch.in.mean = arch.in.mean
    models = 1:11
    names(models) = c("garch", "egarch", "gjr", "aparch", "igarch",
    "figarch.bbm", "figarch.chung", "fiegarch", "fiaparch.bbm",
    "fiaparch.chung", "hygarch")
    selected = strsplit(as.character(formula.var), "\\(")[[2]][1]
    fit$model = models[selected]
    nt = length(series)
    ident = paste(selected, as.character(floor(runif(1) * 10000)),
    sep = "")
    parameters = c(csts = fit$csts, distri = distri, arma = fit$arma.orders,
    arfima = fit$arfima, garch = fit$garch.orders, model = fit$model,
    inmean = fit$arch.in.mean, trunc = truncation, nt = nt)
    write(x = parameters, file = "OxParameter.txt")
    write(x = "X", file = "OxSeries.csv", ncolumns = 1)
    write(x=series, file = "OxSeries.csv", ncolumns = 1, append = TRUE)
    command = "C:\\Ox\\bin\\oxl.exe C:\\Ox\\lib\\GarchOxModelling_w.ox"
    fit$ox = system(command, show.output.on.console = trace,
    invisible = TRUE)
    fit$model = selected
    fit$call = match.call()
    fit$residuals = scan("OxResiduals.csv", skip = 1, quiet = TRUE)
    fit$condvars = scan("OxCondVars.csv", skip = 1, quiet = TRUE)
    fit$coef = matrix(scan("OxParameters.csv", skip = 1, quiet = TRUE),
    byrow = TRUE, ncol = 3)
    fit$title = title
    if (is.null(title))
    fit$title = "GARCH Ox Modelling"
    fit$description = as.character(date())
    class(fit) = "garchOx"
    fit
    }

    You might also like