vis: vis

In LewisLabUCSD/RegressionModelPipeline: Pipeline for feature selection, interaction assessment, regression modeling and prediction assessment for 1-1000s of features

Description

A visualizaiton for univariate and multivariate regressions showing coefficient, confidence intervals and p-values

Usage

vis(out, family = "gaussian", Pr = NULL, fullUnivariate = FALSE,
  intercept = TRUE, col_scheme = rainbow(4), ...)

Arguments

out	a length 2 list univariate and final models: list(screen=several_univariate_models,final=one_multivariate_model). Either screen or final may be set to NULL. Accepts: lm, glm, lmer, glmer
family	a character indicating the glm family fit for the model. If 'binomial' vis_logit will be run. Otherwise standard output is produced.
Pr	an optional numeric vector of the same length and order as out$screen. If the user wants to specify univariate p-values they may do so here. Otherwise the wald test for the 1st variable in the model will be used.
fullUnivariate	visualizes all variables in each univariate model
intercept	boolean indicating if the intercept should be visualized. True by default.
col_scheme	a grDevices::Palettes to specify the color scheme representing -log10(p-values). rainbow(4) by default.

Value

a list of 2 ggplot2 objects visualizing the univariate screening and multivariate final model.

LewisLabUCSD/RegressionModelPipeline documentation built on Jan. 11, 2021, 10:33 p.m.