Cross-Validated Pairs Plot (Helper Functions)

Compute pairwise nonparametric regressions and densities for a set of variables, then plot a pairs-style display with fitted smoothers.

Usage

np.pairs(y_vars, y_dat, ...)
np.pairs.plot(pair_list)

Arguments

Data, Bandwidth Inputs And Formula Interface

These arguments identify the variables, data, and pair specifications to plot.

pair_list: list returned by np.pairs.
y_dat: data frame containing the variables listed in y_vars.
y_vars: character vector of column names in y_dat. If y_vars is named, the names are used as plot labels.

Additional Arguments

Further graphical arguments are passed through to plotting methods.

...: additional arguments passed to npudens and npreg.

Value

np.pairs returns a list with components y_vars, pair_names, and pair_kerns. np.pairs.plot returns NULL (invisibly).

Details

Documentation guide: see np.kernels for kernels, np.options for global options, and plot for plotting options.

On the diagonal, npudens is used to compute kernel density estimates. Off-diagonal panels use npreg with residuals to draw scatterplots and smoothers.

Examples

if (FALSE) { # \dontrun{
data("USArrests")
y_vars <- c("Murder", "UrbanPop")
names(y_vars) <- c("Murder Arrests per 100K", "Pop. Percent Urban")

pair_list <- np.pairs(
  y_vars = y_vars,
  y_dat = USArrests,
  ckertype = "epanechnikov",
  bwscaling = TRUE
)

np.pairs.plot(pair_list)
} # }