Test="ad", test="cvm", test="lillie", or test="skew",Īnd you are testing for some form of normality (i.e., Normal, Lognormal, The default value is =NULL so that all default values for theĮstimating function are used. When test="proucl.ad.gamma" or test="proucl.ks.gamma", you must setĭistribution="gamma" or distribution="gammaAlt".Ī list of arguments to be passed to the function estimating the distribution parameters.įor example, if test="sw" and distribution="gamma", settingĮst.arg.list=list(method="bcmle") indicates using the bias-correctedĮstimators of shape and scale (see the help file for egamma).Įstimating Distribution Parameters for a list of estimating functions. When test="ws", this argument is ignored. When test="chisq", any distribuiton is allowed. When test="ks", any continuous distribution is allowed. "zmlnormAlt" (zero-modified lognormal with alternative parameterization) are allowed. "zmnorm" (zero-modified normal), "zmlnorm" (zero-modified lognormal (delta)), and "lnormAlt" (lognormal with alternative parameterization), Only the values "norm" (normal), "lnorm" (lognormal), When test="ad", test="cvm", test="lillie", or test="skew", "zmlnormAlt" (zero-modified lognormal with alternative parameterization)). (i.e., "zmnorm" (zero-modified normal), "zmlnorm" (zero-modified lognormal (delta)), "gamma" (gamma), etc.), as well as mixed distributions involving the normal distribution When test="sw", test="sf", or test="ppcc", any continuousĭistribuiton is allowed (e.g., "norm" (normal), "lnorm" (lognormal), The default value is distribution="norm" ( Normal distribution). See the help file forĭistribution.df for a list of distributions and their abbreviations.
When the argument x is supplied, you must set test="ks", which is what gofTestĪ character string denoting the distribution abbreviation. Kolmogorov-Smirnov test for a gamma distribution using Anderson-Darling test for a gamma distribution using Wilk-Shapiro test for Uniform distribution. Kolmogorov-Smirnov the default when x IS supplied. Probability Plot Correlation Coefficient. Shapiro-Wilk the default when x is NOT supplied. Inf) values are allowed but will be removed.Ĭharacter string defining which goodness-of-fit test to perform. Missing ( NA), undefined ( NaN), and infinite ( Inf, Kolmogorov-Smirnov goodness-of-fit test ( test="ks"). Numeric vector of values for the first sample in the case of a two-sample Specifies a function which indicates what should happen when the data contain NAs. Specifies an optional vector specifying a subset of observations to be used. If not found in data, the variables are taken fromĮnvironment(formula), typically the environment from which Specifies an optional data frame, list or environment (or object coercibleīy as.ame to a data frame) containing the variables in the This is not a requirement of the test and you can use vectors of differentĪnd infinite ( Inf, -Inf) values are allowed but will be Note thatįor the formula method, x and y must be the same length but Observations in the vector x as the first sample. Observations in the vector y as the second sample and use the Kolmogorov-Smirnov test ( test="ks") and indicates use the Y ~ x is only relevant to the case of the two-sample The vector y for a one-sample goodness-of-fit test. The form y ~ 1 indicates use the observations in In the formula method, y must be a formula of the form y ~ 1 Method, the argument y must be numeric vector of observations. Options$digits,Įxact = NULL, ws.method = "normal scores", warn = TRUE, keep.data = TRUE,ĭata.name = NULL, = NULL, parent.of.data = NULL,Īn object containing data for the goodness-of-fit test. N.param.est = NULL, correct = NULL, digits =. Perform a goodness-of-fit test to determine whether a data setĪppears to come from a specified probability distribution or if twoĭata sets appear to come from the same distribution.ĭistribution = "norm", = NULL,Īlternative = "two.sided", n.classes = NULL,Įstimate.params = ifelse(is.null(param.list), TRUE, FALSE),
0 kommentar(er)
