diag.hgld: Diagnostic for the Hurdle Generalized Lambda Distribution
In dmarcondes/HGLD: Fit the Hurdle Generalized Lambda Distribution

Description Usage Arguments Details Value References Examples

Diagnostic plots and measures of goodness-of-fit for a Hurdle Generalized Lambda Distribution.

diag.hgld(fit, facet = FALSE, facet.breaks, facet.labels, facet.ncol,
  trace = TRUE, no.test = 1000, len = floor(0.9 *
  length(fit$data[fit$data != 0])), alpha = 0.05, plotKS = TRUE,
  KS = TRUE)

`fit`	An object of class fit.hgld.
`facet`	Whether the plots must be faceted for better visualization.
`facet.breaks`	The breaks in which to facet the data. Must be the endpoints of the intervals.
`facet.labels`	The labels of the categories given by the facet breaks.
`facet.ncol`	The number of columns for the facet plot.
`trace`	Whether a progress bar must be printed in order to trace the algorithm.
`no.test`	Total number of KS tests required.
`len`	Number of data to sample at each KS test.
`alpha`	Significance level of KS test.
`plotKS`	Whether to plot the KS resample test result within each plot.
`KS`	Whether the resample KS test must be performed to the non-zero values.

The diagnostics techniques are applied to the non-zero data values. Returns the qq-plot and the quantile plot between the data and the theoretical fitted HGLD. Also returns a table comparing sample moments with theoretical moments. A Kolmogorov-Simornov resample test is performed and the percentage of the times that the null hypotheses, i.e., goodness-of-fit, is not rejected is displayed. All diagnostics are performed for both the RS and fmkl HGLD.

`qqRS`	ggplot qq-plot for the RS HGLD.
`qqfmkl`	ggplot qq-plot for the fmkl HGLD.
`quantRS`	ggplot quantile plot for the RS HGLD.
`quantfmkl`	ggplot quantile plot for the fmkl HGLD.
`moments`	Moments comparison for the GLD fitted to the non-zero data for both parametrizations. These are not the moments of the HGLD, but are instead the moments of the GLD fitted to the non-zero data values.
`KS`	Percentage of no rejection for the KS resample test.

Marcondes, D.; Peixoto, C.; Maia, A. C.; A Survey of a Hurdle Model for Heavy-Tailed Data Based on the Generalized Lambda Distribution. (2017) arxiv1712.02183

Su, S. Fitting Single and Mixture of Generalized Lambda Distributions to Data via Discretized and Maximum Likelihood Methods: GLDEX in R. (2007), Journal of Statistical Software: *21* 9.

set.seed(100)
data <- healthcare[sample(1:nrow(healthcare),50),]
fit <- fit.hgld(data$log_expense)
d <- diag.hgld(fit,facet = FALSE,plotKS = FALSE)

#mixture
set.seed(100)
data <- c(rcauchy(20,location = 10),rep(0,10),rcauchy(20))
fit <- fit.hgld(data = data,mixture = TRUE)
d <- suppressWarnings(diag.hgld(fit))