BTLLasso (version 0.1-5)

singlepaths: Plot parameter paths for BTLLasso

Description

Plots single paths for every parameter of a BTLLasso object or a cv.BTLLasso object. In contrast, to paths, one plot per covariate is created, every single parameter is illustrated by one path. For cv.BTLLasso object, the optimal model according to the cross-validation is marked by a vertical dashed line.

Usage

singlepaths(model, rescale = FALSE, colors = NULL, equal.ranges = FALSE,
  plot.X = TRUE, plot.Z1 = TRUE, plot.Z2 = TRUE, plot.intercepts = TRUE,
  plot.order.effects = TRUE, x.axis = c("loglambda", "lambda", "penalty"),
  columns = NULL, subs.X = NULL, subs.Z1 = NULL, subs.Z2 = NULL)

Arguments

model

BTLLasso or cv.BTLLasso object

rescale

Should the parameter estimates be rescaled for plotting? Only applies if scale = TRUE was specified in BTLLasso or cv.BTLLasso.

colors

Optional. If specified, vector with length equal to the number of objects. Each object can be represented by another color.

equal.ranges

Should all single plots (for different covariates) have equal ranges on the y-axes. FALSE by default.

plot.X

Should paths for variables in X (if present) be plotted?

plot.Z1

Should paths for variables in Z1 (if present) be plotted?

plot.Z2

Should paths for variables in Z2 (if present) be plotted?

plot.intercepts

Should paths for intercepts be plotted separately?

plot.order.effects

Should paths for order effects be plotted separately?

x.axis

Should the paths be plotted against log(lambda+1), against lambda or against the scaled penalty term (between 0 and 1)?

columns

How many columns should be used to arrange the plots. If unspecified, plots are arranged automatically in a quadratic manner.

subs.X

Optional vector of subtitles for variables in X. Can be used to note the encoding of the single covariates, especially for dummy variables.

subs.Z1

Optional vector of subtitles for variables in Z1. Can be used to note the encoding of the single covariates, especially for dummy variables.

subs.Z2

Optional vector of subtitles for variables in Z2. Can be used to note the encoding of the single covariates, especially for dummy variables.

Details

Plots for BTLLasso and cv.BTLLasso objects only differ by the additional vertical line indicating the optimal model according to cross-validation.

References

Schauberger, Gunther and Tutz, Gerhard (2015): Modelling Heterogeneity in Paired Comparison Data - an L1 Penalty Approach with an Application to Party Preference Data, Department of Statistics, LMU Munich, Technical Report 183

Schauberger, Gunther, Groll Andreas and Tutz, Gerhard (2016): Modelling Football Results in the German Bundesliga Using Match-specific Covariates, Department of Statistics, LMU Munich, Technical Report 197

See Also

BTLLasso, cv.BTLLasso, paths

Examples

## Not run: ------------------------------------
# ##### Example with simulated data set containing X, Z1 and Z2
# data(SimData)
# 
# ## Specify tuning parameters
# lambda <- exp(seq(log(151), log(1.05), length = 30)) - 1
# 
# ## Specify control argument
# ## -> allow for object-specific order effects and penalize intercepts
# ctrl <- ctrl.BTLLasso(penalize.intercepts = TRUE, object.order.effect = TRUE,
#                       penalize.order.effect.diffs = TRUE)
# 
# ## Simple BTLLasso model for tuning parameters lambda
# m.sim <- BTLLasso(Y = SimData$Y, X = SimData$X, Z1 = SimData$Z1, 
#                   Z2 = SimData$Z2, lambda = lambda, control = ctrl)
# print(m.sim)
# 
# singlepaths(m.sim)
# 
# ## Cross-validate BTLLasso model for tuning parameters lambda
# set.seed(5)
# m.sim.cv <- cv.BTLLasso(Y = SimData$Y, X = SimData$X, Z1 = SimData$Z1, 
#                         Z2 = SimData$Z2, lambda = lambda, control = ctrl)
# print(m.sim.cv)
# 
# singlepaths(m.sim.cv, plot.order.effect = FALSE, 
#             plot.intercepts = FALSE, plot.Z2 = FALSE)
# paths(m.sim.cv, y.axis = 'L2')
# 
# ## Example for bootstrap confidence intervals for illustration only
# ## Don't calculate bootstrap confidence intervals with B = 10!!!!
# set.seed(5)
# m.sim.boot <- boot.BTLLasso(m.sim.cv, B = 10, cores = 10)
# print(m.sim.boot)
# ci.BTLLasso(m.sim.boot)
# 
# 
# ##### Example with small version from GLES data set
# data(GLESsmall)
# 
# ## extract data and center covariates for better interpretability
# Y <- GLESsmall$Y
# X <- scale(GLESsmall$X, scale = FALSE)
# Z1 <- scale(GLESsmall$Z1, scale = FALSE)
# 
# ## vector of subtitles, containing the coding of the X covariates
# subs.X <- c('', 'female (1); male (0)')
# 
# ## vector of tuning parameters
# lambda <- exp(seq(log(61), log(1), length = 30)) - 1
# 
# 
# ## compute BTLLasso model 
# m.gles <- BTLLasso(Y = Y, X = X, Z1 = Z1, lambda = lambda)
# print(m.gles)
# 
# singlepaths(m.gles, subs.X = subs.X)
# paths(m.gles, y.axis = 'L2')
# 
# ## Cross-validate BTLLasso model 
# m.gles.cv <- cv.BTLLasso(Y = Y, X = X, Z1 = Z1, lambda = lambda)
# print(m.gles.cv)
# 
# singlepaths(m.gles.cv, subs.X = subs.X)
## ---------------------------------------------