Skip to contents

Predicted line from distribution mixture model fit

Source: R/stat-distrmix-line.R
stat_distrmix_line.Rd

stat_distrmix_line() fits a Normal mixture model, by default with normalmixEM(). Predicted values are computed and, by default, plotted.

Usage

stat_distrmix_line(
  mapping = NULL,
  data = NULL,
  geom = "line",
  position = "identity",
  ...,
  orientation = "x",
  method = "normalmixEM",
  se = NULL,
  fit.seed = NA,
  fm.values = FALSE,
  n = min(100 + 50 * k, 300),
  fullrange = TRUE,
  level = 0.95,
  method.args = list(),
  k = 2,
  free.mean = TRUE,
  free.sd = TRUE,
  components = "all",
  n.min = 10L * k,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

mapping

The aesthetic mapping, usually constructed with aes. Only needs to be set at the layer level if you are overriding the plot defaults.

data

A layer specific dataset, only needed if you want to override the plot defaults.

geom

The geometric object to use display the data

position

The position adjustment to use for overlapping points on this layer.

...

other arguments passed on to layer. This can include aesthetics whose values you want to set, not map. See layer for more details.

orientation

character Either "x" or "y", the mapping of the values to which the mixture model is to be fitetd. NOT YET IMPLEMENTED!

method

function or character If character, "normalmixEM" or the name of a model fit function are accepted, possibly followed by the fit function's method argument separated by a colon. The function must return a model fit object of class mixEM.

se

Currently ignored.

fit.seed

RNG seed argument passed to set.seed(). Defaults to NA, which means that set.seed() will not be called.

fm.values

logical Add parameter estimates and their standard errors to the returned values (`FALSE` by default.)

n

Number of points at which to evaluate the model prediction.

fullrange

Should the prediction span the combined range of the scale and of the fitted distributions, or just span the range of the data?

level

Level of confidence interval to use (0.95 by default).

method.args

named list with additional arguments.

k

integer Number of mixture components to fit.

free.mean, free.sd

logical If TRUE, allow the fitted mean and/or fitted sd to vary among the component Normal distributions.

components

character One of "all", "sum", or "members" select which densities are returned.

n.min

integer Minimum number of distinct values in the mapped variable for fitting to the attempted.

na.rm

a logical indicating whether NA values should be stripped before the computation proceeds.

show.legend

logical. Should this layer be included in the legends? NA, the default, includes if any aesthetics are mapped. FALSE never includes, and TRUE always includes.

inherit.aes

If FALSE, overrides the default aesthetics, rather than combining with them. This is most useful for helper functions that define both data and aesthetics and shouldn't inherit behaviour from the default plot specification, e.g. borders.

Value

The value returned by the statistic is a data frame, with n rows of predicted density for each component of the mixture plus their sum and the corresponding vector of x values. Optionally it will also include additional values related to the model fit.

Details

This statistic is similar to stat_density but instead of fitting a single distribution it can fit a mixture of two or more Normal distributions, using an approach related to clustering. Defaults are consistent between stat_distrmix_line() and stat_distrmix_eq(). Parameter fit.seed if not NA is used in a call to set.seed() immediately before calling the model fit function. As the fitting procedure makes use of the (pseudo-)random number generator (RNG), convergence can depend on it, and in such cases setting fit.seed to the same value in stat_distrmix_line() and in stat_distrmix_eq() can ensure consistency, and more generally, reproducibility.

A mixture model as described above, is fitted for k >= 2, while k == 1 is treated as a special case and a Normal distribution fitted with function fitdistr(). In this case the SE values are exact estimates.

Computed variables

stat_distrmix_line() provides the following variables, some of which depend on the orientation:

x

the n values for the quantiles

component

A factor indexing the components and/or their sum

If fm.values = TRUE is passed then columns with diagnosis and parameters estimates are added, with the same value in each row within a group:

n

numeric the number of x values

.size

numeric the number of density values

fm.class

character the most derived class of the fitted model object

fm.method

character the method, as given by the ft field of the fitted model objects

This is wasteful and disabled by default, but provides a simple and robust approach to achieve effects like colouring or hiding of the model fit line by group depending on the outcome of model fitting.

See also

Other ggplot statistics for mixture model fits.: stat_distrmix_eq()

Aesthetics

stat_distrmix_line() understands the following aesthetics. Required aesthetics are displayed in bold and defaults are displayed for optional aesthetics:

•x or y
•group→ after_stat(component)

Learn more about setting these aesthetics in vignette("ggplot2-specs").

Examples

ggplot(faithful, aes(x = waiting)) +
  stat_distrmix_line()


# ggplot(faithful, aes(y = waiting)) +
#  stat_distrmix_line(orientation = "y")

ggplot(faithful, aes(x = waiting)) +
  stat_distrmix_line(components = "sum")


ggplot(faithful, aes(x = waiting)) +
  stat_distrmix_line(components = "members")


ggplot(faithful, aes(x = waiting)) +
 geom_histogram(aes(y = after_stat(density)), bins = 20) +
 stat_distrmix_line(aes(colour = after_stat(component),
                         fill = after_stat(component)),
                     geom = "area", linewidth = 1, alpha = 0.25, se = FALSE)


ggplot(faithful, aes(x = waiting)) +
 stat_distrmix_line(aes(colour = after_stat(component),
                         fill = after_stat(component)),
                     geom = "area", linewidth = 1, alpha = 0.25,
                     components = "members", se = FALSE)


ggplot(faithful, aes(x = waiting)) +
 stat_distrmix_line(geom = "area", linewidth = 1, alpha = 0.25,
                     colour = "black", outline.type = "upper",
                     components = "sum", se = FALSE)


# special case of no mixture
ggplot(subset(faithful, waiting > 66), aes(x = waiting)) +
  stat_distrmix_line(k = 1)
#> With k = 1 one Normal distribution is fitted. Irrelevant parameters ignored!


# Inspecting the returned data using geom_debug_group()
gginnards.installed <- requireNamespace("gginnards", quietly = TRUE)

if (gginnards.installed)
  library(gginnards)

if (gginnards.installed)
  ggplot(faithful, aes(x = waiting)) +
    stat_distrmix_line(geom = "debug_group", components = "all")

#> [1] "PANEL 1; group(s) comp.1; 'draw_function()' input 'data' (head):"
#>           x component      density flipped_aes PANEL  group            y
#> 2  35.29541    comp.1 0.0001092383       FALSE     1 comp.1 0.0001092383
#> 5  35.61754    comp.1 0.0001306554       FALSE     1 comp.1 0.0001306554
#> 8  35.93967    comp.1 0.0001558017       FALSE     1 comp.1 0.0001558017
#> 11 36.26179    comp.1 0.0001852294       FALSE     1 comp.1 0.0001852294
#> 14 36.58392    comp.1 0.0002195534       FALSE     1 comp.1 0.0002195534
#> 17 36.90605    comp.1 0.0002594556       FALSE     1 comp.1 0.0002594556
#>    orientation
#> 2            x
#> 5            x
#> 8            x
#> 11           x
#> 14           x
#> 17           x
#> [1] "PANEL 1; group(s) comp.2; 'draw_function()' input 'data' (head):"
#>           x component      density flipped_aes PANEL  group            y
#> 3  35.29541    comp.2 9.599747e-15       FALSE     1 comp.2 9.599747e-15
#> 6  35.61754    comp.2 1.457545e-14       FALSE     1 comp.2 1.457545e-14
#> 9  35.93967    comp.2 2.206355e-14       FALSE     1 comp.2 2.206355e-14
#> 12 36.26179    comp.2 3.329812e-14       FALSE     1 comp.2 3.329812e-14
#> 15 36.58392    comp.2 5.010203e-14       FALSE     1 comp.2 5.010203e-14
#> 18 36.90605    comp.2 7.515916e-14       FALSE     1 comp.2 7.515916e-14
#>    orientation
#> 3            x
#> 6            x
#> 9            x
#> 12           x
#> 15           x
#> 18           x
#> [1] "PANEL 1; group(s) comp.sum; 'draw_function()' input 'data' (head):"
#>           x component      density flipped_aes PANEL    group            y
#> 1  35.29541  comp.sum 0.0001092383       FALSE     1 comp.sum 0.0001092383
#> 4  35.61754  comp.sum 0.0001306554       FALSE     1 comp.sum 0.0001306554
#> 7  35.93967  comp.sum 0.0001558017       FALSE     1 comp.sum 0.0001558017
#> 10 36.26179  comp.sum 0.0001852294       FALSE     1 comp.sum 0.0001852294
#> 13 36.58392  comp.sum 0.0002195534       FALSE     1 comp.sum 0.0002195534
#> 16 36.90605  comp.sum 0.0002594556       FALSE     1 comp.sum 0.0002594556
#>    orientation
#> 1            x
#> 4            x
#> 7            x
#> 10           x
#> 13           x
#> 16           x

if (gginnards.installed)
  ggplot(faithful, aes(x = waiting)) +
    stat_distrmix_line(geom = "debug_group", components = "sum")

#> [1] "PANEL 1; group(s) comp.sum; 'draw_function()' input 'data' (head):"
#>          x component      density flipped_aes PANEL    group            y
#> 1 35.29551  comp.sum 0.0001092387       FALSE     1 comp.sum 0.0001092387
#> 2 35.61763  comp.sum 0.0001306561       FALSE     1 comp.sum 0.0001306561
#> 3 35.93976  comp.sum 0.0001558028       FALSE     1 comp.sum 0.0001558028
#> 4 36.26189  comp.sum 0.0001852309       FALSE     1 comp.sum 0.0001852309
#> 5 36.58402  comp.sum 0.0002195555       FALSE     1 comp.sum 0.0002195555
#> 6 36.90615  comp.sum 0.0002594584       FALSE     1 comp.sum 0.0002594584
#>   orientation
#> 1           x
#> 2           x
#> 3           x
#> 4           x
#> 5           x
#> 6           x

if (gginnards.installed)
  ggplot(faithful, aes(x = waiting)) +
    stat_distrmix_line(geom = "debug_group", components = "members")

#> [1] "PANEL 1; group(s) comp.1; 'draw_function()' input 'data' (head):"
#>           x component      density flipped_aes PANEL  group            y
#> 1  35.29540    comp.1 0.0001092383       FALSE     1 comp.1 0.0001092383
#> 3  35.61753    comp.1 0.0001306553       FALSE     1 comp.1 0.0001306553
#> 5  35.93966    comp.1 0.0001558016       FALSE     1 comp.1 0.0001558016
#> 7  36.26179    comp.1 0.0001852292       FALSE     1 comp.1 0.0001852292
#> 9  36.58392    comp.1 0.0002195532       FALSE     1 comp.1 0.0002195532
#> 11 36.90604    comp.1 0.0002594554       FALSE     1 comp.1 0.0002594554
#>    orientation
#> 1            x
#> 3            x
#> 5            x
#> 7            x
#> 9            x
#> 11           x
#> [1] "PANEL 1; group(s) comp.2; 'draw_function()' input 'data' (head):"
#>           x component      density flipped_aes PANEL  group            y
#> 2  35.29540    comp.2 9.599325e-15       FALSE     1 comp.2 9.599325e-15
#> 4  35.61753    comp.2 1.457482e-14       FALSE     1 comp.2 1.457482e-14
#> 6  35.93966    comp.2 2.206260e-14       FALSE     1 comp.2 2.206260e-14
#> 8  36.26179    comp.2 3.329671e-14       FALSE     1 comp.2 3.329671e-14
#> 10 36.58392    comp.2 5.009993e-14       FALSE     1 comp.2 5.009993e-14
#> 12 36.90604    comp.2 7.515606e-14       FALSE     1 comp.2 7.515606e-14
#>    orientation
#> 2            x
#> 4            x
#> 6            x
#> 8            x
#> 10           x
#> 12           x

if (gginnards.installed)
  ggplot(faithful, aes(x = waiting)) +
    stat_distrmix_line(geom = "debug_group", fm.values = TRUE)

#> [1] "PANEL 1; group(s) comp.1; 'draw_function()' input 'data' (head):"
#>           x component      density converged   n .size fm.class   fm.method
#> 2  35.29539    comp.1 0.0001092382      TRUE 272   600    mixEM normalmixEM
#> 5  35.61752    comp.1 0.0001306552      TRUE 272   600    mixEM normalmixEM
#> 8  35.93965    comp.1 0.0001558015      TRUE 272   600    mixEM normalmixEM
#> 11 36.26177    comp.1 0.0001852290      TRUE 272   600    mixEM normalmixEM
#> 14 36.58390    comp.1 0.0002195529      TRUE 272   600    mixEM normalmixEM
#> 17 36.90603    comp.1 0.0002594550      TRUE 272   600    mixEM normalmixEM
#>    flipped_aes PANEL  group            y orientation
#> 2        FALSE     1 comp.1 0.0001092382           x
#> 5        FALSE     1 comp.1 0.0001306552           x
#> 8        FALSE     1 comp.1 0.0001558015           x
#> 11       FALSE     1 comp.1 0.0001852290           x
#> 14       FALSE     1 comp.1 0.0002195529           x
#> 17       FALSE     1 comp.1 0.0002594550           x
#> [1] "PANEL 1; group(s) comp.2; 'draw_function()' input 'data' (head):"
#>           x component      density converged   n .size fm.class   fm.method
#> 3  35.29539    comp.2 9.598574e-15      TRUE 272   600    mixEM normalmixEM
#> 6  35.61752    comp.2 1.457369e-14      TRUE 272   600    mixEM normalmixEM
#> 9  35.93965    comp.2 2.206092e-14      TRUE 272   600    mixEM normalmixEM
#> 12 36.26177    comp.2 3.329420e-14      TRUE 272   600    mixEM normalmixEM
#> 15 36.58390    comp.2 5.009620e-14      TRUE 272   600    mixEM normalmixEM
#> 18 36.90603    comp.2 7.515053e-14      TRUE 272   600    mixEM normalmixEM
#>    flipped_aes PANEL  group            y orientation
#> 3        FALSE     1 comp.2 9.598574e-15           x
#> 6        FALSE     1 comp.2 1.457369e-14           x
#> 9        FALSE     1 comp.2 2.206092e-14           x
#> 12       FALSE     1 comp.2 3.329420e-14           x
#> 15       FALSE     1 comp.2 5.009620e-14           x
#> 18       FALSE     1 comp.2 7.515053e-14           x
#> [1] "PANEL 1; group(s) comp.sum; 'draw_function()' input 'data' (head):"
#>           x component      density converged   n .size fm.class   fm.method
#> 1  35.29539  comp.sum 0.0001092382      TRUE 272   600    mixEM normalmixEM
#> 4  35.61752  comp.sum 0.0001306552      TRUE 272   600    mixEM normalmixEM
#> 7  35.93965  comp.sum 0.0001558015      TRUE 272   600    mixEM normalmixEM
#> 10 36.26177  comp.sum 0.0001852290      TRUE 272   600    mixEM normalmixEM
#> 13 36.58390  comp.sum 0.0002195529      TRUE 272   600    mixEM normalmixEM
#> 16 36.90603  comp.sum 0.0002594550      TRUE 272   600    mixEM normalmixEM
#>    flipped_aes PANEL    group            y orientation
#> 1        FALSE     1 comp.sum 0.0001092382           x
#> 4        FALSE     1 comp.sum 0.0001306552           x
#> 7        FALSE     1 comp.sum 0.0001558015           x
#> 10       FALSE     1 comp.sum 0.0001852290           x
#> 13       FALSE     1 comp.sum 0.0002195529           x
#> 16       FALSE     1 comp.sum 0.0002594550           x