| w.starplot.Coxmos {Coxmos} | R Documentation |
w.starplot.Coxmos
Description
The w.starplot.Coxmos function offers a graphical representation of the W* (W star) values from
a given Coxmos model. Through this visualization, users can gain insights into the variable
contributions and their significance in the model. The function provides options for customization,
allowing users to focus on specific variables, exclude zero values, and adjust the visual limits.
Usage
w.starplot.Coxmos(model, zero.rm = FALSE, top = NULL, auto.limits = TRUE)
Arguments
model |
Coxmos model. |
zero.rm |
Logical. Remove variables equal to 0 (default: FALSE). |
top |
Numeric. Show "top" first variables. If top = NULL, all variables are shown (default: NULL). |
auto.limits |
Logical. If "auto.limits" = TRUE, limits are detected automatically (default: TRUE). |
Details
The w.starplot.Coxmos function is tailored to visualize the W* values, which are indicative of
the variable contributions in a Coxmos model. Initially, the function checks the class of the
provided model to ensure its compatibility with the Coxmos framework.
The W* values are extracted from the model and subsequently processed based on user-defined
parameters. The zero.rm option allows users to exclude variables with zero W* values, ensuring
a more concise visualization. If the top parameter is specified, the function focuses on
displaying only the top-ranked variables based on their absolute W* values.
The visualization is constructed using the 'ggplot2' framework. The color scale can be automatically
adjusted to the maximum absolute W* value when the auto.limits parameter is set to TRUE. The
function also checks for the availability of the RColorConesa package. If present, it leverages
its color palettes for a more refined visualization; in its absence, default color schemes are applied.
Value
A list of ggplot2 objects, each representing the W* values for a component of
the Coxmos model.
Examples
data("X_proteomic")
data("Y_proteomic")
X <- X_proteomic[,1:50]
Y <- Y_proteomic
splsicox.model <- splsicox(X, Y, n.comp = 2, penalty = 0.5, x.center = TRUE, x.scale = TRUE)
w.starplot.Coxmos(model = splsicox.model)