mclustDensity estimationDiagnostic plots for density estimation. Only available for the one-dimensional case.
densityMclust.diagnostic(object, type = c("cdf", "qq"),
col = c("black", "black"),
lwd = c(2,1), lty = c(1,1),
legend = TRUE, grid = TRUE,
…)An object of class 'mclustDensity' obtained from a call to densityMclust function.
The type of graph requested:
"cdf" =a plot of the estimated CDF versus the empirical distribution function.
"qq" =a Q-Q plot of sample quantiles versus the quantiles obtained from the inverse of the estimated cdf.
A pair of values for the color to be used for plotting, respectively, the estimated CDF and the empirical cdf.
A pair of values for the line width to be used for plotting, respectively, the estimated CDF and the empirical cdf.
A pair of values for the line type to be used for plotting, respectively, the estimated CDF and the empirical cdf.
A logical indicating if a legend must be added to the plot of fitted CDF vs the empirical CDF.
A logical indicating if a grid should be added to the plot.
Additional arguments.
The two diagnostic plots for density estimation in the one-dimensional case are discussed in Loader (1999, pp- 87-90).
Loader C. (1999), Local Regression and Likelihood. New York, Springer.
C. Fraley, A. E. Raftery, T. B. Murphy and L. Scrucca (2012). mclust Version 4 for R: Normal Mixture Modeling for Model-Based Clustering, Classification, and Density Estimation. Technical Report No. 597, Department of Statistics, University of Washington.
# NOT RUN {
x <- faithful$waiting
dens <- densityMclust(x)
plot(dens, x, what = "diagnostic")
# or
densityMclust.diagnostic(dens, type = "cdf")
densityMclust.diagnostic(dens, type = "qq")
# }
Run the code above in your browser using DataLab