smoothtab: Presmoothing and postsmoothing of empirical distributions

In twolodzko/equi: equi: Equipercentile Equating

Description

Function for Log-linear presmoothing and/or Gaussian Kernel postsmoothing.

Usage

smoothtab(x, y, presmoothing = FALSE, postsmoothing = FALSE,
  bandwidth = "auto", lldeg = 4, llxdeg = 1, raw = TRUE, cdf = TRUE,
  margin = 0.5, grid = 100)

as.smoothtab(x)

is.smoothtab(x)

## S3 method for class 'smoothtab'
plot(x, type = "s", lty = 1:6, add = FALSE, ...)

## S3 method for class 'smoothtab'
cdfplot(x, add = FALSE, ...)

Arguments

x,	y numeric vectors.
presmoothing	if TRUE Log-linear presmoothing is applied.
postsmoothing	if TRUE Gaussian Kernel postsmoothing is applied.
bandwidth	sets bandwidth for Kernel Smoothing. Use "auto" (default) for automatic selection of bandwidth.
lldeg	degree of the polynomial in log-linear presmoothing (deafult is 4).
llxdeg	degree of the polynomial in log-linear presmoothing for interaction term (deafult is 1).
raw	if TRUE computes raw polynomials for log-linear presmoothing, see: poly.
cdf	if FALSE compute probabilities rather than cumulative probabilities (default).
margin	if postsmoothing=TRUE, it defines the margins of points range to be created using Gaussian Kernel postsmoothing. The function returns point values +/- margin.
grid	if postsmoothing=TRUE, it defines the number of points to be created using Gaussian Kernel postsmoothing.
type	type of the plot.
lty	a vector of line types, see: par.
add	add a plot to previous one.
...	potentially further arguments passed from other methods.

Value

Returns two-column data.frame with unique score points and coresponding probabilities, or a list of two data.frames for marginal probabilities of joint distributions.

Note

See also equate and kequate packages.

References

Holland, P.W. & Thayer, D.T. (2000). Univariate and Bivariate Loglinear Models for Discrete Test Score Distributions. Journal of Educational and Behavioral Statistics, 25(2), 133-183.

Kolen, M.J. & Brennan, R.J. (2004). Test Equating, Scaling, and Linking: Methods and Practices. New York: Springer-Verlag.

von Davier, A.A., Holland, P.W. & Thayer, D.T. (2004). The Kernel Method of Test Equating. New York: Springer-Verlag.

Wand, M.P. & Jones, M.C. (1995). Kernel Smoothing. London: Chapman & Hall/CRC.

Examples

data(Tests)
x <- Tests[Tests$Sample == "P", "x"]
y <- Tests[Tests$Sample == "P", "y"]

(st <- smoothtab(x))
plot(st)

smoothtab(x, presmoothing=TRUE)
smoothtab(x, y, presmoothing=TRUE, postsmoothing=TRUE)

twolodzko/equi documentation built on Nov. 11, 2020, 4:04 p.m.